Linden Street Clinic / A / A3 dimensioned submission

2 authored page(s) + 6 required evidence page(s) · revision 2

Document UC-MED-A-submission · source src-facbccd0-f1d3-4b37-9e10-c511f73c228f · 2026-10-09T12:17:24.272Z

{"contentHash":"fnv1a-502ba6e6","documentId":"UC-MED-A-submission","rendererVersion":"laydyne-report-svg-1.0.0","revision":2,"sourceId":"src-facbccd0-f1d3-4b37-9e10-c511f73c228f"}{"frame":{"height":13,"width":396,"x":12,"y":10},"id":"dimensioned-plan-title","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":18,"weight":"bold"},"text":"Linden Street Clinic / A / shared reception and payment / dimensioned submission","type":"text"}Linden Street Clinic / A / shared reception and payment / dimensioned submission{"frame":{"height":10,"width":396,"x":12,"y":27},"id":"dimensioned-plan-status","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":10,"weight":"normal"},"text":"FOR COMPARISON ONLY — assumed concept dimensions, not surveyed or issued for construction. Current process engine 2.7.0; no historical output reused.","type":"text"}FOR COMPARISON ONLY — assumed concept dimensions, not surveyed or issued for construction. Current process engine 2.7.0; no historical output reused.{"caption":"A / shared reception and payment. 21 × 15 m shell. Page inset: use written dimensions, not its printed scale. Fixed walls/doors, four rooms, passages and medical parts match across A/B.","comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","dimensions":true,"floorId":"floor-1","frame":{"height":214,"width":271,"x":12,"y":45},"id":"dimensioned-plan-plan","model":"comparison","side":"baseline","type":"drawing"}Floor plan of L1 · Linden Street Clinic / UC-MED-01 A — shared reception at 1:150 on A4 landscape: 21.00 × 15.00 m, 188 parts, 24 zones (18 named, 6 numbered), floor area 315.00 m². Inside: 11 aisle widths (narrowest 1.20 m) and 15 zone sides dimensioned where they fit.Linden Street Clinic / UC-MED-01 A — shared reception — L1 · Linden Street Clinic / UC-MED-01 A — shared reception — floor planLaydyne drawing. Scale 1:150 on A4 landscape; units metres; space fnv1a-11feab13; 2026-10-09.21.00 m15.00 m1.20 m1.40 m1.50 m1.50 m1.55 m1.75 m2.25 m2.80 m2.80 m2.95 m2.95 m3.60 m3.60 m3.60 m3.60 m6.15 m4.50 m4.00 m2.40 m2.40 m3.00 m4.00 m8.05 m3.55 m4.20 m3.50 mWaiting · 20 seats + 2 wheelchair spaces28.58 m²Staff room / lockers / kitchenette27.68 m²Clinical circulation · 1.65 m min main lane27.14 m²X-ray · 4.0 × 4.518.00 m²Treatment · 4.0 × 4.016.00 m²Consult 4 · 3.0 × 3.610.80 m²Consult 3 · 3.0 × 3.610.80 m²Consult 1 · 3.0 × 3.610.80 m²Consult 2 · 3.0 × 3.610.80 m²Patient circulation8.88 m²Kids corner8.25 m²X-ray approach · 1.5 m clear8.02 m²Reception / payment · 2 staffNurse / clean utility7.20 m²Linen / storage · 1.7 × 3.66.12 m²WC · 1.8 × 2.44.32 m²X-ray control4.18 m²WheelchairKids corner rugReception / paymentKitchenetteNurse worktopStaff tableWC1989222324Areas (m²)1Consult 1 · 3.0 × 3.610.802Consult 2 · 3.0 × 3.610.803Consult 3 · 3.0 × 3.610.804Consult 4 · 3.0 × 3.610.805Linen / storage · 1.7 × 3.66.126Staff room / lockers / kitchenette27.687Clinical circulation · 1.65 m min main lane27.148Accessible WC · 2.4 × 2.45.769Ø 1.5 m clear turning circle1.7610WC · 1.8 × 2.44.3211Nurse / clean utility7.2012X-ray control4.1813Treatment · 4.0 × 4.016.0014Patient circulation8.8815X-ray · 4.0 × 4.518.0016X-ray approach · 1.5 m clear8.0217Reception / payment · 2 staff7.3518Waiting · 20 seats + 2 wheelchair spaces28.5819Wind lobby · wheelchair / buggy parking11.7620Kids corner8.2521Wheelchair1.2622Wheelchair1.2623Wheelchair park1.1924Buggy park1.04Floor area315.00Zoned (overlaps once)231.22Not zoned83.78Seats (chairs, stools)41Fixtures54 Cabinet8 Washbasin8 Locker7Zone overlap in rows7.7302.55m1:150Linden Street Clinic / UC-MED-01 A — shared receptionL1 · Linden Street Clinic / UC-MED-01 A — shared reception — floor planDimensions not yet measured on site. Areas from the drawn outlines, less holes.Scale 1:150 · A4 landscapeUnits: metres · 2026-10-09Space fnv1a-11feab13LaydyneA / shared reception and payment. 21 × 15 m shell. Page inset: use written dimensions, not its printed scale. Fixed walls/doors, four rooms, passages and medical parts match across A/B.Source src-facbccd0-f1d3-4b37-9e10-c511f73c228f · fnv1a-11feab13 · 2026-10-09 · e1c7afaf-93af-4d9c-baea-350c4f28811b/baselineDrawing inset: use written dimensions; the original sheet scale does not apply.{"frame":{"height":34,"width":111,"x":297,"y":45},"id":"dimensioned-plan-fixed","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":10,"weight":"normal"},"text":"315 m² shell. Four consultation rooms: 3.0 × 3.6 m each. 20 waiting chairs plus two modeled wheelchair bays. Staff total seven in each option; B splits reception/payment one + one.","type":"text"}315 m² shell. Four consultation rooms: 3.0 × 3.6 m each. 20 waiting chairs plus two modeled wheelchair bays. Staff total seven in each option; B splits reception/payment one + one.{"borderColor":"#bbc4cc","caption":"Geometric side gaps from inherited footprints; they are not medical or supplier-required clearances.","columns":[{"align":"left","id":"part","title":"Medical part","width":34},{"align":"left","id":"point","title":"Assumed side / x,z (m)","width":52},{"align":"left","id":"gap","title":"Side gap","width":25}],"frame":{"height":103,"width":111,"x":297,"y":86},"headerColor":"#edf1f4","id":"dimensioned-plan-equipment-access","rows":[{"cells":{"gap":{"kind":"text","text":"0.95 m"},"part":{"kind":"text","text":"Consult 1 couch"},"point":{"kind":"text","text":"west / 2.025, 1.15"}},"id":"c1"},{"cells":{"gap":{"kind":"text","text":"0.95 m"},"part":{"kind":"text","text":"Consult 2 couch"},"point":{"kind":"text","text":"west / 5.175, 1.15"}},"id":"c2"},{"cells":{"gap":{"kind":"text","text":"0.95 m"},"part":{"kind":"text","text":"Consult 3 couch"},"point":{"kind":"text","text":"west / 8.325, 1.15"}},"id":"c3"},{"cells":{"gap":{"kind":"text","text":"0.95 m"},"part":{"kind":"text","text":"Consult 4 couch"},"point":{"kind":"text","text":"west / 11.475, 1.15"}},"id":"c4"},{"cells":{"gap":{"kind":"text","text":"1.20 m"},"part":{"kind":"text","text":"Left treatment couch"},"point":{"kind":"text","text":"east / 1.55, 10.05"}},"id":"t0"},{"cells":{"gap":{"kind":"text","text":"1.24 m"},"part":{"kind":"text","text":"Right treatment couch"},"point":{"kind":"text","text":"west / 2.85, 10.05"}},"id":"t1"},{"cells":{"gap":{"kind":"text","text":"1.75 m"},"part":{"kind":"text","text":"X-ray unit"},"point":{"kind":"text","text":"south / 19.40, 9.90"}},"id":"xr"}],"style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.12,"size":8.5,"weight":"normal"},"type":"table"}Medical partAssumed side / x,z (m)Side gapConsult 1 couchwest / 2.025, 1.150.95 mConsult 2 couchwest / 5.175, 1.150.95 mConsult 3 couchwest / 8.325, 1.150.95 mConsult 4 couchwest / 11.475, 1.150.95 mLeft treatment coucheast / 1.55, 10.051.20 mRight treatment couchwest / 2.85, 10.051.24 mX-ray unitsouth / 19.40, 9.901.75 mGeometric side gaps from inherited footprints; they are not medical or supplier-required clearances.{"frame":{"height":39,"width":111,"x":297,"y":197},"id":"dimensioned-plan-reach","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":9,"weight":"normal"},"text":"Patient circle r=0.30 m and r=0.45 m; staff r=0.30 m. A: 49 targets; B: 51, all reachable, including the seven explicit operation-side points. Staff origin 15.5,4.25 m; patient origin 10.5,14.1 m.","type":"text"}Patient circle r=0.30 m and r=0.45 m; staff r=0.30 m. A: 49 targets; B: 51, all reachable, including the seven explicit operation-side points. Staff origin 15.5,4.25 m; patient origin 10.5,14.1 m.{"frame":{"height":17,"width":271,"x":12,"y":267},"id":"dimensioned-plan-passages","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":9,"weight":"normal"},"text":"Passage-tag zones checked at 1.50 m; global minimum at 0.90 m. Consultation doors are modeled 1.00 m clear, treatment and X-ray 1.20 m. Door width is not proof of a 1.50 m passage.","type":"text"}Passage-tag zones checked at 1.50 m; global minimum at 0.90 m. Consultation doors are modeled 1.00 m clear, treatment and X-ray 1.20 m. Door width is not proof of a 1.50 m passage.{"frame":{"height":41,"width":111,"x":297,"y":241},"id":"dimensioned-plan-unverified","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":9,"weight":"normal"},"text":"Unverified: real wheelchair/bed turning, transfers, door construction, supplier envelopes, clinical safety and shielding. Retractable-curtain operation is not simulated.","type":"text"}Unverified: real wheelchair/bed turning, transfers, door construction, supplier envelopes, clinical safety and shielding. Retractable-curtain operation is not simulated.
{"contentHash":"fnv1a-502ba6e6","documentId":"UC-MED-A-submission","rendererVersion":"laydyne-report-svg-1.0.0","revision":2,"sourceId":"src-facbccd0-f1d3-4b37-9e10-c511f73c228f"}{"frame":{"height":14,"width":396,"x":12,"y":10},"id":"comparison-and-methods-title","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":18,"weight":"bold"},"text":"B shortens movement, increases patient wait, and retains a seat-shortage risk","type":"text"}B shortens movement, increases patient wait, and retains a seat-shortage risk{"frame":{"height":16,"width":396,"x":12,"y":29},"id":"comparison-and-methods-result-scope","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":10,"weight":"normal"},"text":"20 paired seeds, six-hour window, warmup 0. Each option: 867 arrived = 867 completed in total, unfinished 0, blocked 0. All 40 run states completed, with no event/job limit or warning. No seed was omitted.","type":"text"}20 paired seeds, six-hour window, warmup 0. Each option: 867 arrived = 867 completed in total, unfinished 0, blocked 0. All 40 run states completed, with no event/job limit or warning. No seed was omitted.{"borderColor":"#bbc4cc","caption":"Means of the twenty run-level measures. All main metrics use all twenty pairs. An interval containing zero means these runs do not establish a difference.","columns":[{"align":"left","id":"metric","title":"Measure","width":67},{"align":"left","id":"A","title":"A","width":36},{"align":"left","id":"B","title":"B","width":36},{"align":"left","id":"delta","title":"B - A / paired 95% interval","width":105}],"frame":{"height":154,"width":244,"x":12,"y":53},"headerColor":"#edf1f4","id":"comparison-and-methods-comparison","rows":[{"cells":{"A":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"baseline"}},"B":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"candidate"}},"delta":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"delta"}},"metric":{"kind":"text","text":"Arrived patients per seed"}},"id":"arrived"},{"cells":{"A":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"baseline"}},"B":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"candidate"}},"delta":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"delta"}},"metric":{"kind":"text","text":"Total patient wait / arrived"}},"id":"wait"},{"cells":{"A":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"baseline"}},"B":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"candidate"}},"delta":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"delta"}},"metric":{"kind":"text","text":"Cycle time / completed"}},"id":"cycle"},{"cells":{"A":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"baseline"}},"B":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"candidate"}},"delta":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"delta"}},"metric":{"kind":"text","text":"Total patient distance / 6 h"}},"id":"distance"},{"cells":{"A":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"baseline"}},"B":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"candidate"}},"delta":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"delta"}},"metric":{"kind":"text","text":"Check-in resource wait"}},"id":"checkin"},{"cells":{"A":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"baseline"}},"B":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"candidate"}},"delta":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"delta"}},"metric":{"kind":"text","text":"First doctor resource wait"}},"id":"doctor"},{"cells":{"A":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"baseline"}},"B":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"candidate"}},"delta":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"delta"}},"metric":{"kind":"text","text":"Payment resource wait"}},"id":"payment"},{"cells":{"A":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"baseline"}},"B":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"candidate"}},"delta":{"decimals":2,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"delta"}},"metric":{"kind":"text","text":"Waiting-demand peak"}},"id":"peak"}],"style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.12,"size":9,"weight":"normal"},"type":"table"}MeasureABB - A / paired 95% intervalArrived patients per seed43.35 count43.35 count0.00 count95% interval: [0.00, 0.00] · includes 0Total patient wait / arrived1,902.16 s1,922.54 s20.37 s95% interval: [9.97, 30.77]Cycle time / completed2,854.29 s2,857.25 s2.95 s95% interval: [-7.45, 13.36] · includes 0Total patient distance / 6 h4,101.45 m3,345.46 m-755.99 m95% interval: [-822.68, -689.30]Check-in resource wait18.71 s122.93 s104.23 s95% interval: [71.32, 137.14]First doctor resource wait1,698.37 s1,619.50 s-78.87 s95% interval: [-109.94, -47.80]Payment resource wait9.48 s11.31 s1.83 s95% interval: [-0.71, 4.37] · includes 0Waiting-demand peak14.30 count13.75 count-0.55 count95% interval: [-0.91, -0.19]Means of the twenty run-level measures. All main metrics use all twenty pairs. An interval containing zero means these runs do not establish a difference.{"borderColor":"#bbc4cc","caption":"Ratio 0.50 = 50%. Held unit-seconds / (capacity × 21600 s), including travel and other waits while held; not clinical active work. Split pools have no cross-ID paired delta.","columns":[{"align":"left","id":"resource","title":"Resource","width":58},{"align":"left","id":"A","title":"A / held share","width":39},{"align":"left","id":"B","title":"B / held share","width":39}],"frame":{"height":150,"width":136,"x":272,"y":53},"headerColor":"#edf1f4","id":"comparison-and-methods-utilization","rows":[{"cells":{"A":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"receptionists","scope":"resource","side":"baseline"}},"B":{"kind":"text","text":"Separate pools below"},"resource":{"kind":"text","text":"Shared reception (2)"}},"id":"receptionists"},{"cells":{"A":{"kind":"text","text":"Shared pool above"},"B":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"checkin-receptionist","scope":"resource","side":"candidate"}},"resource":{"kind":"text","text":"Check-in (1)"}},"id":"checkin-receptionist"},{"cells":{"A":{"kind":"text","text":"Shared pool above"},"B":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"payment-receptionist","scope":"resource","side":"candidate"}},"resource":{"kind":"text","text":"Payment (1)"}},"id":"payment-receptionist"},{"cells":{"A":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"baseline"}},"B":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"candidate"}},"resource":{"kind":"text","text":"Doctors (2)"}},"id":"doctors"},{"cells":{"A":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"baseline"}},"B":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"candidate"}},"resource":{"kind":"text","text":"Nurses (2)"}},"id":"nurses"},{"cells":{"A":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"baseline"}},"B":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"candidate"}},"resource":{"kind":"text","text":"Radiographer (1)"}},"id":"radiographer"},{"cells":{"A":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"baseline"}},"B":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"candidate"}},"resource":{"kind":"text","text":"Consult room 1"}},"id":"room1"},{"cells":{"A":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"baseline"}},"B":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"candidate"}},"resource":{"kind":"text","text":"Consult room 2"}},"id":"room2"},{"cells":{"A":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"baseline"}},"B":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"candidate"}},"resource":{"kind":"text","text":"Consult room 3"}},"id":"room3"},{"cells":{"A":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"baseline"}},"B":{"decimals":4,"kind":"metric","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"candidate"}},"resource":{"kind":"text","text":"Consult room 4"}},"id":"room4"}],"style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.12,"size":8.5,"weight":"normal"},"type":"table"}ResourceA / held shareB / held shareShared reception (2)0.2128 ratiomissing_valueSeparate pools belowCheck-in (1)Shared pool above0.2443 ratiomissing_valuePayment (1)Shared pool above0.1812 ratiomissing_valueDoctors (2)0.5057 ratio0.5057 ratioNurses (2)0.1219 ratio0.1219 ratioRadiographer (1)0.0871 ratio0.0871 ratioConsult room 10.2650 ratio0.2654 ratioConsult room 20.2565 ratio0.2582 ratioConsult room 30.2446 ratio0.2446 ratioConsult room 40.2452 ratio0.2432 ratioRatio 0.50 = 50%. Held unit-seconds / (capacity × 21600 s), including travel and other waits while held; not clinical active work. Split pools have no cross-ID paired delta.{"frame":{"height":62,"width":244,"x":12,"y":219},"id":"comparison-and-methods-definitions","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":9.5,"weight":"normal"},"text":"Patient wait includes every arrived patient and any unfinished waits up to the endpoint. Here unfinished is zero in every seed. Cycle means use completed patients only. First-doctor/check-in/payment rows are means per request to those steps; the total patient wait also includes return consultation and other resources. Return-doctor wait: A 1757.55 s, B 1682.57 s, each valid in 19/20 seeds; seed 20261027 has no imaging patient. No full-20 paired delta is calculated for this missing measure.\nDistance per arrived patient is calculated within each seed, then averaged: A 94.76 m; B 77.34 m. Paired difference -17.42 m, 95% [-17.55, -17.29].\nSame arrival ID/order, patient service draws and medical-branch order are retained. The two nurses serve one representative treatment point; bed assignment and staff motion are absent. B changes geometry and staff allocation together; isolate those changes in a later controlled candidate before attributing an effect to either alone.","type":"text"}Patient wait includes every arrived patient and any unfinished waits up to the endpoint. Here unfinished is zero in every seed. Cycle means use completed patients only. First-doctor/check-in/payment rows are means per request to those steps; the total patient wait also includes return consultation and other resources. Return-doctor wait: A 1757.55 s, B 1682.57 s, each valid in 19/20 seeds; seed 20261027 has no imaging patient. No full-20 paired delta is calculated for this missing measure.Distance per arrived patient is calculated within each seed, then averaged: A 94.76 m; B 77.34 m. Paired difference -17.42 m, 95% [-17.55, -17.29].Same arrival ID/order, patient service draws and medical-branch order are retained. The two nurses serve one representative treatment point; bed assignment and staff motion are absent. B changes geometry and staff allocation together; isolate those changes in a later controlled candidate before attributing an effect to either alone.{"frame":{"height":65,"width":136,"x":272,"y":215},"id":"comparison-and-methods-seats","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":9,"weight":"normal"},"text":"20 physical seats; waiting-demand capacity 1000 records demand without pushing excess patients into an outside admission queue. Per-seed shortage = max(0, peak − 20): A mean 0.75 / max 12; B mean 0.55 / max 9. Both 3/20 (15%) seeds exceed the seats. Shortage delta −0.20, 95% [−0.53, 0.13], includes zero.\nOccupancy includes entry/exit movement for patients stopping in the waiting polygon; it is an approximation to demand, not seat allocation. First seed only: sampled time above 20 is 0/21600 s in each option, at 180 s intervals. That is not a twenty-seed duration mean.","type":"text"}20 physical seats; waiting-demand capacity 1000 records demand without pushing excess patients into an outside admission queue. Per-seed shortage = max(0, peak − 20): A mean 0.75 / max 12; B mean 0.55 / max 9. Both 3/20 (15%) seeds exceed the seats. Shortage delta −0.20, 95% [−0.53, 0.13], includes zero.Occupancy includes entry/exit movement for patients stopping in the waiting polygon; it is an approximation to demand, not seat allocation. First seed only: sampled time above 20 is 0/21600 s in each option, at 180 s intervals. That is not a twenty-seed duration mean.{"frame":{"height":8,"width":396,"x":12,"y":284},"id":"comparison-and-methods-evidence","style":{"align":"left","color":"#17212b","font":"sans","lineHeight":1.22,"size":7,"weight":"normal"},"text":"Frozen comparison e1c7afaf-93af-4d9c-baea-350c4f28811b · process-des-2.7.0 · source 45f6a99d · Assumed input; field prediction, safety and legal compliance unverified.","type":"text"}Frozen comparison e1c7afaf-93af-4d9c-baea-350c4f28811b · process-des-2.7.0 · source 45f6a99d · Assumed input; field prediction, safety and legal compliance unverified.
{"documentId":"UC-MED-A-submission","metrics":[{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":0,"low":0,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":0,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1902.16270762074,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1922.536314288935,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":30.77435012764803,"low":9.972863208741476,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":20.373606668194753,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":2854.2932563953214,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":2857.2482302390945,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":13.364472367035209,"low":-7.454524679489832,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":2.954973843772689,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"m","value":4101.452453229362,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"m","value":3345.4639028562538,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-689.2960843409162,"low":-822.6810164052984,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"m","value":-755.9885503731073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":18.705477380671418,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":122.93101026935054,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":137.1356328873578,"low":71.31543289000035,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":104.22553288867907,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1698.3705860772639,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1619.499102056592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-47.80229344552859,"low":-109.94067459581501,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":-78.8714840206718,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":9.484422133528332,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":11.312732338517694,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":4.366852259871348,"low":-0.7102318498926252,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":1.8283102049893614,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":14.3,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":13.75,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-0.19470876827297678,"low":-0.9052912317270233,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":-0.55,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"receptionists","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.21275376189386913,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"checkin-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24430847037628864,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"payment-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.18119905341144965,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056519563426073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056665668460949,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2649691770995189,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2654073935533545,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25653721914499594,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25822529104225106,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2446198412868495,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24455027088001763,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24517767515385,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24315017821656623,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]}],"rendererVersion":"laydyne-report-svg-1.0.0","revision":2,"sourceId":"src-facbccd0-f1d3-4b37-9e10-c511f73c228f"}Evidence appendix 1/6Document UC-MED-A-submission · revision 2 · source src-facbccd0-f1d3-4b37-9e10-c511f73c228fMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.arrived / run / run / baselineValue: 43.35 count95% delta interval: not available for this sideBound: none; reason: noneComparison: e1c7afaf-93af-4d9c-baea-350c4f28811bEngine: process-des-2.7.0Baseline input: fnv1a-aff24d3eCandidate input: fnv1a-b0c2596cBaseline space: fnv1a-11feab13Candidate space: fnv1a-9911c034Horizon: 21600 s; warmup: 0 s; selection: completionSeed: 20261009; runs: 20baseline warnings (0):nonebaseline assumptions (17):1. Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.2. Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.3. Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.4. Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.5. Evacuation, safety codes and legal compliance, lines of sight and cameras.6. Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).7. Cleaning quality (coverage comes from simulate_space).8. Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.9. Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.10. Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.11. goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.12. Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.13. Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.14. A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.15. Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.16. Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.17. Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.candidate warnings (0):noneCaptured 2026-10-09T12:14:53.805Z · fnv1a-502ba6e6
{"documentId":"UC-MED-A-submission","metrics":[{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":0,"low":0,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":0,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1902.16270762074,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1922.536314288935,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":30.77435012764803,"low":9.972863208741476,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":20.373606668194753,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":2854.2932563953214,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":2857.2482302390945,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":13.364472367035209,"low":-7.454524679489832,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":2.954973843772689,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"m","value":4101.452453229362,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"m","value":3345.4639028562538,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-689.2960843409162,"low":-822.6810164052984,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"m","value":-755.9885503731073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":18.705477380671418,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":122.93101026935054,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":137.1356328873578,"low":71.31543289000035,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":104.22553288867907,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1698.3705860772639,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1619.499102056592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-47.80229344552859,"low":-109.94067459581501,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":-78.8714840206718,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":9.484422133528332,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":11.312732338517694,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":4.366852259871348,"low":-0.7102318498926252,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":1.8283102049893614,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":14.3,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":13.75,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-0.19470876827297678,"low":-0.9052912317270233,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":-0.55,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"receptionists","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.21275376189386913,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"checkin-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24430847037628864,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"payment-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.18119905341144965,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056519563426073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056665668460949,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2649691770995189,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2654073935533545,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25653721914499594,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25822529104225106,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2446198412868495,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24455027088001763,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24517767515385,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24315017821656623,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]}],"rendererVersion":"laydyne-report-svg-1.0.0","revision":2,"sourceId":"src-facbccd0-f1d3-4b37-9e10-c511f73c228f"}Evidence appendix 2/6Document UC-MED-A-submission · revision 2 · source src-facbccd0-f1d3-4b37-9e10-c511f73c228fcandidate assumptions (17):1. Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.2. Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.3. Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.4. Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.5. Evacuation, safety codes and legal compliance, lines of sight and cameras.6. Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).7. Cleaning quality (coverage comes from simulate_space).8. Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.9. Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.10. Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.11. goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.12. Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.13. Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.14. A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.15. Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.16. Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.17. Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.Different arrivals: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.arrived / run / run / candidateValue: 43.35 count95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.arrived / run / run / deltaValue: 0 count95% delta interval: [0, 0]; includes 0: trueBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.wait.mean / run / run / baselineValue: 1902.16270762074 s95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.wait.mean / run / run / candidateCaptured 2026-10-09T12:14:53.805Z · fnv1a-502ba6e6
{"documentId":"UC-MED-A-submission","metrics":[{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":0,"low":0,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":0,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1902.16270762074,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1922.536314288935,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":30.77435012764803,"low":9.972863208741476,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":20.373606668194753,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":2854.2932563953214,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":2857.2482302390945,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":13.364472367035209,"low":-7.454524679489832,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":2.954973843772689,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"m","value":4101.452453229362,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"m","value":3345.4639028562538,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-689.2960843409162,"low":-822.6810164052984,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"m","value":-755.9885503731073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":18.705477380671418,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":122.93101026935054,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":137.1356328873578,"low":71.31543289000035,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":104.22553288867907,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1698.3705860772639,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1619.499102056592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-47.80229344552859,"low":-109.94067459581501,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":-78.8714840206718,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":9.484422133528332,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":11.312732338517694,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":4.366852259871348,"low":-0.7102318498926252,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":1.8283102049893614,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":14.3,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":13.75,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-0.19470876827297678,"low":-0.9052912317270233,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":-0.55,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"receptionists","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.21275376189386913,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"checkin-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24430847037628864,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"payment-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.18119905341144965,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056519563426073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056665668460949,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2649691770995189,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2654073935533545,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25653721914499594,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25822529104225106,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2446198412868495,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24455027088001763,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24517767515385,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24315017821656623,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]}],"rendererVersion":"laydyne-report-svg-1.0.0","revision":2,"sourceId":"src-facbccd0-f1d3-4b37-9e10-c511f73c228f"}Evidence appendix 3/6Document UC-MED-A-submission · revision 2 · source src-facbccd0-f1d3-4b37-9e10-c511f73c228fValue: 1922.536314288935 s95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.wait.mean / run / run / deltaValue: 20.373606668194753 s95% delta interval: [9.972863208741476, 30.77435012764803]; includes 0: falseBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.cycle.mean / run / run / baselineValue: 2854.2932563953214 s95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.cycle.mean / run / run / candidateValue: 2857.2482302390945 s95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.cycle.mean / run / run / deltaValue: 2.954973843772689 s95% delta interval: [-7.454524679489832, 13.364472367035209]; includes 0: trueBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.distance.total / run / run / baselineValue: 4101.452453229362 m95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.distance.total / run / run / candidateValue: 3345.4639028562538 m95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / run.distance.total / run / run / deltaValue: -755.9885503731073 m95% delta interval: [-822.6810164052984, -689.2960843409162]; includes 0: falseBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / process.time.mean / process / checkin-wait / baselineValue: 18.705477380671418 s95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / process.time.mean / process / checkin-wait / candidateValue: 122.93101026935054 s95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / process.time.mean / process / checkin-wait / deltaValue: 104.22553288867907 s95% delta interval: [71.31543289000035, 137.1356328873578]; includes 0: falseBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / process.time.mean / process / doctor-wait / baselineValue: 1698.3705860772639 sCaptured 2026-10-09T12:14:53.805Z · fnv1a-502ba6e6
{"documentId":"UC-MED-A-submission","metrics":[{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":0,"low":0,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":0,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1902.16270762074,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1922.536314288935,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":30.77435012764803,"low":9.972863208741476,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":20.373606668194753,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":2854.2932563953214,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":2857.2482302390945,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":13.364472367035209,"low":-7.454524679489832,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":2.954973843772689,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"m","value":4101.452453229362,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"m","value":3345.4639028562538,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-689.2960843409162,"low":-822.6810164052984,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"m","value":-755.9885503731073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":18.705477380671418,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":122.93101026935054,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":137.1356328873578,"low":71.31543289000035,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":104.22553288867907,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1698.3705860772639,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1619.499102056592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-47.80229344552859,"low":-109.94067459581501,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":-78.8714840206718,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":9.484422133528332,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":11.312732338517694,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":4.366852259871348,"low":-0.7102318498926252,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":1.8283102049893614,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":14.3,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":13.75,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-0.19470876827297678,"low":-0.9052912317270233,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":-0.55,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"receptionists","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.21275376189386913,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"checkin-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24430847037628864,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"payment-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.18119905341144965,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056519563426073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056665668460949,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2649691770995189,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2654073935533545,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25653721914499594,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25822529104225106,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2446198412868495,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24455027088001763,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24517767515385,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24315017821656623,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]}],"rendererVersion":"laydyne-report-svg-1.0.0","revision":2,"sourceId":"src-facbccd0-f1d3-4b37-9e10-c511f73c228f"}Evidence appendix 4/6Document UC-MED-A-submission · revision 2 · source src-facbccd0-f1d3-4b37-9e10-c511f73c228f95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / process.time.mean / process / doctor-wait / candidateValue: 1619.499102056592 s95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / process.time.mean / process / doctor-wait / deltaValue: -78.8714840206718 s95% delta interval: [-109.94067459581501, -47.80229344552859]; includes 0: falseBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / process.time.mean / process / payment-wait / baselineValue: 9.484422133528332 s95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / process.time.mean / process / payment-wait / candidateValue: 11.312732338517694 s95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / process.time.mean / process / payment-wait / deltaValue: 1.8283102049893614 s95% delta interval: [-0.7102318498926252, 4.366852259871348]; includes 0: trueBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / zone.occupancy.peak / zone / waiting-demand / baselineValue: 14.3 count95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / zone.occupancy.peak / zone / waiting-demand / candidateValue: 13.75 count95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / zone.occupancy.peak / zone / waiting-demand / deltaValue: -0.55 count95% delta interval: [-0.9052912317270233, -0.19470876827297678]; includes 0: falseBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / receptionists / baselineValue: 0.21275376189386913 ratio95% delta interval: not available for this sideBound: none; reason: missing_valueMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / checkin-receptionist / candidateValue: 0.24430847037628864 ratio95% delta interval: not available for this sideBound: none; reason: missing_valueMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / payment-receptionist / candidateValue: 0.18119905341144965 ratio95% delta interval: not available for this sideCaptured 2026-10-09T12:14:53.805Z · fnv1a-502ba6e6
{"documentId":"UC-MED-A-submission","metrics":[{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":0,"low":0,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":0,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1902.16270762074,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1922.536314288935,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":30.77435012764803,"low":9.972863208741476,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":20.373606668194753,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":2854.2932563953214,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":2857.2482302390945,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":13.364472367035209,"low":-7.454524679489832,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":2.954973843772689,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"m","value":4101.452453229362,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"m","value":3345.4639028562538,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-689.2960843409162,"low":-822.6810164052984,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"m","value":-755.9885503731073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":18.705477380671418,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":122.93101026935054,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":137.1356328873578,"low":71.31543289000035,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":104.22553288867907,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1698.3705860772639,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1619.499102056592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-47.80229344552859,"low":-109.94067459581501,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":-78.8714840206718,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":9.484422133528332,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":11.312732338517694,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":4.366852259871348,"low":-0.7102318498926252,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":1.8283102049893614,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":14.3,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":13.75,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-0.19470876827297678,"low":-0.9052912317270233,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":-0.55,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"receptionists","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.21275376189386913,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"checkin-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24430847037628864,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"payment-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.18119905341144965,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056519563426073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056665668460949,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2649691770995189,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2654073935533545,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25653721914499594,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25822529104225106,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2446198412868495,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24455027088001763,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24517767515385,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24315017821656623,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]}],"rendererVersion":"laydyne-report-svg-1.0.0","revision":2,"sourceId":"src-facbccd0-f1d3-4b37-9e10-c511f73c228f"}Evidence appendix 5/6Document UC-MED-A-submission · revision 2 · source src-facbccd0-f1d3-4b37-9e10-c511f73c228fBound: none; reason: missing_valueMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / doctors / baselineValue: 0.5056519563426073 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / doctors / candidateValue: 0.5056665668460949 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / nurses / baselineValue: 0.12189310230216592 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / nurses / candidateValue: 0.12189310230216592 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / radiographer / baselineValue: 0.08711011776053709 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / radiographer / candidateValue: 0.08711011776053709 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / room1 / baselineValue: 0.2649691770995189 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / room1 / candidateValue: 0.2654073935533545 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / room2 / baselineValue: 0.25653721914499594 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / room2 / candidateValue: 0.25822529104225106 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / room3 / baselineValue: 0.2446198412868495 ratio95% delta interval: not available for this sideBound: none; reason: noneCaptured 2026-10-09T12:14:53.805Z · fnv1a-502ba6e6
{"documentId":"UC-MED-A-submission","metrics":[{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":0,"low":0,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":0,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1902.16270762074,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1922.536314288935,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":30.77435012764803,"low":9.972863208741476,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":20.373606668194753,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":2854.2932563953214,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":2857.2482302390945,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":13.364472367035209,"low":-7.454524679489832,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":2.954973843772689,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"m","value":4101.452453229362,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"m","value":3345.4639028562538,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-689.2960843409162,"low":-822.6810164052984,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"m","value":-755.9885503731073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":18.705477380671418,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":122.93101026935054,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":137.1356328873578,"low":71.31543289000035,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":104.22553288867907,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1698.3705860772639,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1619.499102056592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-47.80229344552859,"low":-109.94067459581501,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":-78.8714840206718,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":9.484422133528332,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":11.312732338517694,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":4.366852259871348,"low":-0.7102318498926252,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":1.8283102049893614,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":14.3,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":13.75,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-0.19470876827297678,"low":-0.9052912317270233,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":-0.55,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"receptionists","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.21275376189386913,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"checkin-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24430847037628864,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"payment-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.18119905341144965,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056519563426073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056665668460949,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2649691770995189,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2654073935533545,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25653721914499594,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25822529104225106,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2446198412868495,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24455027088001763,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24517767515385,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24315017821656623,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]}],"rendererVersion":"laydyne-report-svg-1.0.0","revision":2,"sourceId":"src-facbccd0-f1d3-4b37-9e10-c511f73c228f"}Evidence appendix 6/6Document UC-MED-A-submission · revision 2 · source src-facbccd0-f1d3-4b37-9e10-c511f73c228fMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / room3 / candidateValue: 0.24455027088001763 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / room4 / baselineValue: 0.24517767515385 ratio95% delta interval: not available for this sideBound: none; reason: noneMetric: e1c7afaf-93af-4d9c-baea-350c4f28811b / resource.utilization / resource / room4 / candidateValue: 0.24315017821656623 ratio95% delta interval: not available for this sideBound: none; reason: noneCaptured 2026-10-09T12:14:53.805Z · fnv1a-502ba6e6
Frozen source and complete recorded evidence
{"documentId":"UC-MED-A-submission","metrics":[{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":43.35,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":0,"low":0,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.arrived","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":0,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1902.16270762074,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1922.536314288935,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":30.77435012764803,"low":9.972863208741476,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.wait.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":20.373606668194753,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":2854.2932563953214,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":2857.2482302390945,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":13.364472367035209,"low":-7.454524679489832,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.cycle.mean","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":2.954973843772689,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"baseline"},"runs":20,"seed":20261009,"unit":"m","value":4101.452453229362,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"candidate"},"runs":20,"seed":20261009,"unit":"m","value":3345.4639028562538,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-689.2960843409162,"low":-822.6810164052984,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"run.distance.total","resourceId":null,"scope":"run","side":"delta"},"runs":20,"seed":20261009,"unit":"m","value":-755.9885503731073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":18.705477380671418,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":122.93101026935054,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":137.1356328873578,"low":71.31543289000035,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"checkin-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":104.22553288867907,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":1698.3705860772639,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":1619.499102056592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-47.80229344552859,"low":-109.94067459581501,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"doctor-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":-78.8714840206718,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"baseline"},"runs":20,"seed":20261009,"unit":"s","value":9.484422133528332,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"candidate"},"runs":20,"seed":20261009,"unit":"s","value":11.312732338517694,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":4.366852259871348,"low":-0.7102318498926252,"withinNoise":true},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"process.time.mean","resourceId":"payment-wait","scope":"process","side":"delta"},"runs":20,"seed":20261009,"unit":"s","value":1.8283102049893614,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"baseline"},"runs":20,"seed":20261009,"unit":"count","value":14.3,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"candidate"},"runs":20,"seed":20261009,"unit":"count","value":13.75,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":{"high":-0.19470876827297678,"low":-0.9052912317270233,"withinNoise":false},"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"zone.occupancy.peak","resourceId":"waiting-demand","scope":"zone","side":"delta"},"runs":20,"seed":20261009,"unit":"count","value":-0.55,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"receptionists","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.21275376189386913,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"checkin-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24430847037628864,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":"missing_value","ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"payment-receptionist","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.18119905341144965,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056519563426073,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"doctors","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.5056665668460949,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"nurses","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.12189310230216592,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"radiographer","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.08711011776053709,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2649691770995189,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room1","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2654073935533545,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25653721914499594,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room2","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.25822529104225106,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.2446198412868495,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room3","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24455027088001763,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"baseline"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24517767515385,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]},{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation.","Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"baselineInputHash":"fnv1a-aff24d3e","baselineSpaceHash":"fnv1a-11feab13","bound":null,"candidateInputHash":"fnv1a-b0c2596c","candidateSpaceHash":"fnv1a-9911c034","differentArrivals":null,"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"interval":null,"reason":null,"ref":{"comparisonId":"e1c7afaf-93af-4d9c-baea-350c4f28811b","metricId":"resource.utilization","resourceId":"room4","scope":"resource","side":"candidate"},"runs":20,"seed":20261009,"unit":"ratio","value":0.24315017821656623,"warmupSeconds":0,"warmupSelect":"completion","warnings":[]}],"rendererVersion":"laydyne-report-svg-1.0.0","revision":2,"source":{"building":{"activeFloor":"floor-1","floors":[{"id":"floor-1","level":0,"scene":{"assumptions":["Agent-created spatial draft. Verify object footprints and passage widths.","Robot settings, workload and station positions are illustrative defaults."],"coverageWidth":0.6,"demandPerHour":30,"depth":15,"destination":{"x":10.5,"z":6},"dock":{"x":10.5,"z":14.1},"durationMinutes":30,"fleetSize":1,"kind":"facility","name":"Linden Street Clinic / UC-MED-01 A — shared reception","objects":[{"color":14410462,"d":0.2,"h":3,"id":"north","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":21,"x":0,"z":0},{"color":14410462,"d":14.6,"h":3,"id":"west","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.2,"x":0,"z":0.2},{"color":14410462,"d":14.6,"h":3,"id":"east","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.2,"x":20.8,"z":0.2},{"color":14410462,"d":0.2,"h":3,"id":"south","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":21,"x":0,"z":14.8},{"color":11919333,"d":0.2,"h":2.9,"id":"glaze-w","kind":"window","label":"Street glazing","note":"Full-height shopfront assumed; mullions to survey.","source":"assumed","w":9.4,"x":0.2,"y":0.05,"z":14.8},{"color":11919333,"d":0.2,"h":2.9,"id":"glaze-e","kind":"window","label":"Street glazing","note":"Full-height shopfront assumed; mullions to survey.","source":"assumed","w":9.4,"x":11.4,"y":0.05,"z":14.8},{"d":0.2,"doorType":"automatic","h":2.2,"id":"entrance","kind":"door","label":"Entrance 1.8 m automatic","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.8,"x":9.6,"z":14.8},{"d":0.2,"doorType":"hinged","h":2.2,"id":"service","kind":"door","label":"Staff / service exit 1.1 m","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.1,"x":19.1,"z":0},{"color":6911869,"d":0.5,"h":3,"id":"col-7-5","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":6.75,"z":4.75},{"color":6911869,"d":0.5,"h":3,"id":"col-7-10","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":6.75,"z":9.75},{"color":6911869,"d":0.5,"h":3,"id":"col-14-5","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":13.75,"z":4.75},{"color":6911869,"d":0.5,"h":3,"id":"col-14-10","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":13.75,"z":9.75},{"color":14410462,"d":3.6,"h":3,"id":"c1-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":3.2,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c1-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":0.2,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c1","kind":"zone","label":"Consult 1 · 3.0 × 3.6","source":"assumed","w":3,"x":0.2,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c1-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":0.35,"z":3.8},{"d":0.6,"h":0.75,"id":"c1-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":0.4,"z":0.95},{"d":0.5,"h":0.85,"id":"c1-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":0.65,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c1-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":0.65,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c1-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":2.5,"z":0.2},{"d":0.45,"h":0.85,"id":"c1-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":2.6,"z":3.15},{"color":14410462,"d":3.6,"h":3,"id":"c2-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":6.35,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c2-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":3.35,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c2","kind":"zone","label":"Consult 2 · 3.0 × 3.6","source":"assumed","w":3,"x":3.35,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c2-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":3.5,"z":3.8},{"d":0.6,"h":0.75,"id":"c2-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":3.5500000000000003,"z":0.95},{"d":0.5,"h":0.85,"id":"c2-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":3.8000000000000003,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c2-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":3.8,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c2-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":5.6499999999999995,"z":0.2},{"d":0.45,"h":0.85,"id":"c2-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":5.75,"z":3.15},{"color":14410462,"d":3.6,"h":3,"id":"c3-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":9.5,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c3-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":6.5,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c3","kind":"zone","label":"Consult 3 · 3.0 × 3.6","source":"assumed","w":3,"x":6.5,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c3-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":7.55,"z":3.8},{"d":0.6,"h":0.75,"id":"c3-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":6.7,"z":0.95},{"d":0.5,"h":0.85,"id":"c3-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":6.95,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c3-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":6.95,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c3-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":8.8,"z":0.2},{"d":0.45,"h":0.85,"id":"c3-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":8.899999999999999,"z":3.15},{"color":14410462,"d":3.6,"h":3,"id":"c4-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":12.649999999999999,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c4-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":9.649999999999999,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c4","kind":"zone","label":"Consult 4 · 3.0 × 3.6","source":"assumed","w":3,"x":9.649999999999999,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c4-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":9.799999999999999,"z":3.8},{"d":0.6,"h":0.75,"id":"c4-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":9.849999999999998,"z":0.95},{"d":0.5,"h":0.85,"id":"c4-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":10.099999999999998,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c4-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":10.1,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c4-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":11.95,"z":0.2},{"d":0.45,"h":0.85,"id":"c4-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":12.049999999999999,"z":3.15},{"color":14410462,"d":4.5,"h":3,"id":"staff-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":14.5,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"staff-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":6.3,"x":14.5,"z":4.7},{"color":15196113,"d":4.5,"h":0.02,"id":"staff","kind":"zone","label":"Staff room / lockers / kitchenette","source":"assumed","w":6.15,"x":14.65,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"staff-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":15,"z":4.7},{"d":0.5,"h":1.8,"id":"locker-0","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":14.85,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-1","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":15.33,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-2","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":15.809999999999999,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-3","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":16.29,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-4","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":16.77,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-5","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":17.25,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-6","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":17.73,"z":0.4},{"d":0.6,"h":0.9,"id":"kitchen","kind":"kitchen","label":"Kitchenette","note":"Client to confirm kitchenette dimensions, specification and placement.","source":"assumed","w":2.4,"x":18.15,"z":4.1},{"d":0.65,"h":1.8,"id":"fridge","kind":"fridge","label":"Fridge","note":"Client to confirm fridge dimensions, specification and placement.","source":"assumed","w":0.65,"x":20.15,"z":3.05},{"d":0.75,"h":0.74,"id":"staff-table","kind":"table","label":"Staff table","note":"Client to confirm staff table dimensions, specification and placement.","source":"assumed","w":1.5,"x":16.65,"z":2.05},{"d":0.5,"h":0.85,"id":"staff-chair-0-0","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":16.65,"z":1.35},{"d":0.5,"facing":180,"h":0.85,"id":"staff-chair-0-1","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":16.65,"z":3.15},{"d":0.5,"h":0.85,"id":"staff-chair-1-0","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":17.5,"z":1.35},{"d":0.5,"facing":180,"h":0.85,"id":"staff-chair-1-1","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":17.5,"z":3.15},{"color":15789020,"d":1.65,"h":0.02,"id":"clinical","kind":"zone","label":"Clinical circulation · 1.65 m min main lane","source":"assumed","tags":["passage"],"w":16.45,"x":0.2,"z":5.25},{"color":15789020,"d":4.8,"h":0.02,"id":"spine","kind":"zone","label":"Patient circulation","source":"assumed","tags":["passage"],"w":1.85,"x":9.45,"z":7.05},{"color":14410462,"d":4,"h":3,"id":"treat-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":4.2,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"treat-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":0.05000000000000002,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"treat-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":0.05000000000000002,"z":11.05},{"color":14018790,"d":4,"h":0.02,"id":"treat","kind":"zone","label":"Treatment · 4.0 × 4.0","source":"assumed","w":4,"x":0.2,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"treat-door","kind":"door","label":"Treatment 1.2 m clear","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.2,"x":2.85,"z":6.9},{"color":12703967,"d":2,"h":0.65,"id":"treat-bed-0","kind":"exam_couch","label":"Treatment couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.75,"x":0.2,"z":9.05},{"color":12703967,"d":2,"h":0.65,"id":"treat-bed-1","kind":"exam_couch","label":"Treatment couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.75,"x":3.45,"z":9.05},{"blocks":false,"d":1.8,"h":0.3,"id":"treat-screen","kind":"partition","label":"Ceiling privacy curtain track","note":"Client to confirm ceiling privacy curtain track dimensions, specification and placement.","source":"assumed","w":0.08,"x":2.15,"y":2.2,"z":9.15},{"d":0.5,"h":0.85,"id":"treat-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.6,"x":0.5,"z":7.3},{"d":0.45,"h":1,"id":"treat-supply","kind":"cabinet","label":"Clinical supplies","note":"Client to confirm clinical supplies dimensions, specification and placement.","source":"assumed","w":1,"x":1.25,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"awc-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":4.35,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"awc-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":6.9,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"awc-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.6999999999999997,"x":4.35,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"awc-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.6999999999999997,"x":4.35,"z":9.45},{"color":14736624,"d":2.4,"h":0.02,"id":"awc","kind":"zone","label":"Accessible WC · 2.4 × 2.4","source":"assumed","w":2.4,"x":4.5,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"awc-door","kind":"door","label":"Accessible WC 1.0 m clear","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":5.5,"z":6.9},{"d":0.7,"facing":180,"h":0.8,"id":"awc-toilet","kind":"toilet","label":"WC with side-transfer space","note":"Client to confirm grab rails: locations, heights, fixings and transfer side with accessibility specialist; retain clear turning area.","source":"assumed","w":0.4,"x":4.75,"z":8.55},{"d":0.45,"h":0.8,"id":"awc-basin","kind":"washbasin","label":"Accessible basin","note":"Client to confirm accessible basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":4.65,"z":7.3},{"color":15854840,"d":1.5,"footprint":[{"x":1.5,"z":0.75},{"x":1.4855889603024228,"z":0.8963177415120962},{"x":1.442909649383465,"z":1.0370125742738172},{"x":1.373602209226909,"z":1.1666776747647016},{"x":1.2803300858899107,"z":1.2803300858899105},{"x":1.1666776747647019,"z":1.373602209226909},{"x":1.0370125742738174,"z":1.442909649383465},{"x":0.8963177415120962,"z":1.4855889603024228},{"x":0.75,"z":1.5},{"x":0.6036822584879038,"z":1.4855889603024228},{"x":0.4629874257261827,"z":1.442909649383465},{"x":0.33332232523529853,"z":1.3736022092269091},{"x":0.2196699141100894,"z":1.2803300858899107},{"x":0.126397790773091,"z":1.1666776747647016},{"x":0.057090350616535,"z":1.0370125742738174},{"x":0.014411039697577177,"z":0.8963177415120964},{"x":0,"z":0.7500000000000001},{"x":0.014411039697577177,"z":0.6036822584879038},{"x":0.05709035061653489,"z":0.46298742572618273},{"x":0.12639779077309088,"z":0.33332232523529853},{"x":0.2196699141100893,"z":0.2196699141100894},{"x":0.33332232523529837,"z":0.1263977907730911},{"x":0.46298742572618223,"z":0.05709035061653511},{"x":0.6036822584879035,"z":0.014411039697577288},{"x":0.7499999999999999,"z":0},{"x":0.8963177415120962,"z":0.014411039697577177},{"x":1.0370125742738174,"z":0.057090350616535},{"x":1.1666776747647014,"z":0.12639779077309088},{"x":1.2803300858899105,"z":0.2196699141100893},{"x":1.373602209226909,"z":0.33332232523529837},{"x":1.442909649383465,"z":0.46298742572618223},{"x":1.4855889603024228,"z":0.6036822584879035}],"h":0.02,"id":"awc-turn","kind":"zone","label":"Ø 1.5 m clear turning circle","source":"assumed","symbol":"wheelchair","w":1.5,"x":5.25,"z":7.65},{"color":14410462,"d":2.4,"h":3,"id":"wc-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":7.35,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"wc-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":9.3,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"wc-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1,"x":7.35,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"wc-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1,"x":7.35,"z":9.45},{"color":14736624,"d":2.4,"h":0.02,"id":"wc","kind":"zone","label":"WC · 1.8 × 2.4","source":"assumed","w":1.8,"x":7.5,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"wc-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":7.65,"z":6.9},{"d":0.7,"facing":180,"h":0.8,"id":"wc-toilet","kind":"toilet","label":"WC","note":"Client to confirm wc dimensions, specification and placement.","source":"assumed","w":0.4,"x":8.55,"z":8.55},{"d":0.4,"h":0.85,"id":"wc-basin","kind":"washbasin","label":"Basin","note":"Client to confirm basin dimensions, specification and placement.","source":"assumed","w":0.45,"x":8.65,"z":7.3},{"color":14410462,"d":2.4,"h":3,"id":"utility-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":11.299999999999999,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"utility-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":14.45,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"utility-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3.3,"x":11.299999999999999,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"utility-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3.3,"x":11.299999999999999,"z":9.45},{"color":14871000,"d":2.4,"h":0.02,"id":"utility","kind":"zone","label":"Nurse / clean utility","source":"assumed","w":3,"x":11.45,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"utility-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":11.8,"z":6.9},{"d":0.55,"h":0.9,"id":"utility-work","kind":"counter","label":"Nurse worktop","note":"Client to confirm nurse worktop dimensions, specification and placement.","source":"assumed","w":2.45,"x":11.7,"z":8.9},{"d":0.45,"h":0.85,"id":"utility-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.55,"x":13.65,"z":7.3},{"color":14410462,"d":2.2,"h":3,"id":"control-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":14.6,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"control-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1999999999999997,"x":14.6,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"control-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1999999999999997,"x":14.6,"z":9.25},{"color":14475241,"d":2.2,"h":0.02,"id":"control","kind":"zone","label":"X-ray control","source":"assumed","w":1.9,"x":14.75,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"control-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":14.9,"z":6.9},{"d":1.3,"h":0.75,"id":"control-console","kind":"desk","label":"Control console","note":"Client to confirm control console dimensions, specification and placement.","source":"assumed","w":0.55,"x":16.1,"z":7.35},{"d":0.5,"facing":90,"h":0.85,"id":"control-chair","kind":"chair","label":"Radiographer","note":"Client to confirm radiographer dimensions, specification and placement.","source":"assumed","w":0.5,"x":15.1,"z":8.35},{"color":14410462,"d":4.5,"h":3,"id":"xray-w","kind":"wall","label":"Control shielding screen · shielding by specialist","note":"shielding by specialist; confirm barrier, glazing, door and equipment operating envelopes.","source":"assumed","w":0.15,"x":16.650000000000002,"z":6.55},{"color":14410462,"d":0.15,"h":3,"id":"xray-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":16.650000000000002,"z":6.3999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"xray-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":16.650000000000002,"z":11.049999999999999},{"color":14475241,"d":4.5,"h":0.02,"id":"xray","kind":"zone","label":"X-ray · 4.0 × 4.5","source":"assumed","w":4,"x":16.8,"z":6.55},{"d":1.2,"doorType":"sliding","h":2.2,"id":"xray-door","kind":"door","label":"X-ray 1.2 m clear","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":0.15,"x":16.65,"z":9.75},{"d":1.1,"h":0.8,"id":"control-window","kind":"window","label":"Specialist shielding / viewing glass","note":"All X-ray shielding, door, glazing and equipment clearances by radiation specialist.","source":"assumed","w":0.15,"x":16.65,"y":1.1,"z":7.45},{"d":1.2,"h":2,"id":"xray-table","kind":"xray_unit","label":"X-ray unit / representative equipment","note":"Current X-ray unit replaces the old table and box stand-ins. Assumed supplier envelope; radiation shielding, clinical use and operating clearances are not validated.","source":"assumed","w":2.4,"x":18.2,"z":8.1},{"d":0.45,"h":1.1,"id":"xray-store","kind":"cabinet","label":"Imaging storage","note":"Client to confirm imaging storage dimensions, specification and placement.","source":"assumed","w":1.1,"x":17.15,"z":6.55},{"color":15067865,"d":3.55,"h":0.02,"id":"waiting","kind":"zone","label":"Waiting · 20 seats + 2 wheelchair spaces","source":"assumed","w":8.05,"x":0.2,"z":11.25},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-0","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":0.5,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-1","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":1.1800000000000002,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-2","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":1.86,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-3","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":2.54,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-4","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.22,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-5","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.9000000000000004,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-6","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":4.58,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-7","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.260000000000001,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-8","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.94,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-9","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":6.62,"z":11.3},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-0","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":2.65,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-1","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.2,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-2","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.75,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-3","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":4.3,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-4","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":4.85,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-5","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.4,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-6","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.95,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-7","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":6.5,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-8","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":7.05,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-9","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":7.6,"z":14.15},{"color":13098711,"d":1.4,"h":0.02,"id":"wheel-0","kind":"zone","label":"Wheelchair","note":"Keep clear. Client to confirm wheelchair bay dimensions and approach.","source":"assumed","symbol":"wheelchair","w":0.9,"x":0.5,"z":13.35},{"color":13098711,"d":1.4,"h":0.02,"id":"wheel-1","kind":"zone","label":"Wheelchair","note":"Keep clear. Client to confirm wheelchair bay dimensions and approach.","source":"assumed","symbol":"wheelchair","w":0.9,"x":1.65,"z":13.35},{"color":14410462,"d":2.8,"h":3,"id":"lobby-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":8.25,"z":12},{"color":14410462,"d":2.8,"h":3,"id":"lobby-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":12.6,"z":12},{"color":14410462,"d":0.15,"h":3,"id":"lobby-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.5,"x":8.25,"z":11.85},{"color":15002091,"d":2.8,"h":0.02,"id":"lobby","kind":"zone","label":"Wind lobby · wheelchair / buggy parking","source":"assumed","w":4.2,"x":8.4,"z":12},{"d":0.15,"doorType":"automatic","h":2.2,"id":"lobby-inner","kind":"door","label":"Lobby 1.8 m automatic","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.8,"x":9.6,"z":11.85},{"color":13098711,"d":1.4,"h":0.02,"id":"lobby-wc","kind":"zone","label":"Wheelchair park","source":"assumed","symbol":"wheelchair","w":0.85,"x":8.55,"z":13.35},{"color":14278618,"d":1.3,"h":0.02,"id":"lobby-buggy","kind":"zone","label":"Buggy park","source":"assumed","w":0.8,"x":11.6,"z":13.35},{"color":15261127,"d":2.1,"h":0.02,"id":"reception","kind":"zone","label":"Reception / payment · 2 staff","source":"assumed","w":3.5,"x":12.9,"z":11.4},{"color":11836537,"d":0.7,"h":1.05,"id":"reception-counter","kind":"counter","label":"Reception / payment","note":"Client to confirm reception / payment dimensions, specification and placement.","source":"assumed","w":2.6,"x":12.75,"z":12.6},{"d":0.7,"h":0.76,"id":"reception-low","kind":"counter","label":"Accessible counter","note":"Lower section; detail knee recess and hearing loop.","source":"assumed","w":1,"x":15.35,"z":12.6},{"d":0.55,"h":0.85,"id":"recept-chair-0","kind":"chair","label":"Receptionist","note":"Client to confirm receptionist dimensions, specification and placement.","source":"assumed","w":0.55,"x":13.35,"z":11.8},{"d":0.55,"h":0.85,"id":"recept-chair-1","kind":"chair","label":"Receptionist","note":"Client to confirm receptionist dimensions, specification and placement.","source":"assumed","w":0.55,"x":14.95,"z":11.8},{"color":15851979,"d":2.5,"h":0.02,"id":"kids","kind":"zone","label":"Kids corner","source":"assumed","w":3.3,"x":17.2,"z":12.1},{"d":0.7,"h":0.5,"id":"kids-table","kind":"round_table","label":"Kids table","note":"Client to confirm kids table dimensions, specification and placement.","source":"assumed","w":0.7,"x":18.4,"z":13},{"d":0.35,"h":0.3,"id":"kids-stool-0","kind":"stool","label":"Kids seat","note":"Client to confirm kids seat dimensions, specification and placement.","source":"assumed","w":0.35,"x":17.9,"z":13.2},{"d":0.35,"h":0.3,"id":"kids-stool-1","kind":"stool","label":"Kids seat","note":"Client to confirm kids seat dimensions, specification and placement.","source":"assumed","w":0.35,"x":19.2,"z":13.2},{"d":1.5,"h":0.8,"id":"kids-shelf","kind":"bookshelf","label":"Books / toys","note":"Client to confirm books / toys dimensions, specification and placement.","source":"assumed","w":0.35,"x":20.45,"z":12.4},{"color":15196113,"d":3.6,"h":0.02,"id":"linen","kind":"zone","label":"Linen / storage · 1.7 × 3.6","source":"assumed","w":1.7,"x":12.8,"z":0.2},{"d":0.15,"h":3,"id":"linen-s","kind":"wall","label":"Storage partition","source":"assumed","w":1.7,"x":12.8,"z":3.8},{"d":0.15,"doorType":"sliding","h":2.2,"id":"linen-door","kind":"door","label":"Storage 1.0 m clear","source":"assumed","w":1,"x":13,"z":3.8},{"d":2.2,"h":1.8,"id":"linen-shelves","kind":"shelf","label":"Linen shelving","note":"Client to confirm linen shelving dimensions, specification and placement.","source":"assumed","w":0.5,"x":14,"z":0.2},{"color":14410462,"d":2.4,"h":3,"id":"toilet-chase","kind":"block","label":"Plumbing service void","source":"assumed","w":0.3,"x":7.05,"z":7.05},{"d":1.55,"h":2.75,"id":"lobby-glass-w","kind":"window","label":"Glazed lobby screen","source":"assumed","w":0.15,"x":8.25,"y":0.1,"z":13.2},{"d":2.7,"h":2.75,"id":"lobby-glass-e","kind":"window","label":"Glazed lobby screen","source":"assumed","w":0.15,"x":12.6,"y":0.1,"z":12.05},{"d":0.15,"h":2.75,"id":"lobby-glass-nw","kind":"window","label":"Lobby glazing","source":"assumed","w":1.2,"x":8.4,"y":0.1,"z":11.85},{"d":0.15,"h":2.75,"id":"lobby-glass-ne","kind":"window","label":"Lobby glazing","source":"assumed","w":1.2,"x":11.4,"y":0.1,"z":11.85},{"d":1.1,"doorType":"sliding","h":2.2,"id":"waiting-door","kind":"door","label":"Waiting 1.1 m clear sliding","source":"assumed","w":0.15,"x":8.25,"z":12.1},{"d":1.5,"h":0.02,"id":"xray-approach","kind":"zone","label":"X-ray approach · 1.5 m clear","source":"assumed","tags":["passage"],"w":5.35,"x":11.3,"z":10.25},{"d":0.5,"facing":90,"h":0.85,"id":"c1-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":0.3,"z":2.37},{"d":0.35,"h":1.1,"id":"c1-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":1.65,"z":0.2},{"d":0.5,"facing":90,"h":0.85,"id":"c2-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":3.45,"z":2.37},{"d":0.35,"h":1.1,"id":"c2-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":4.8,"z":0.2},{"d":0.5,"facing":90,"h":0.85,"id":"c3-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":6.6,"z":2.37},{"d":0.35,"h":1.1,"id":"c3-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":7.95,"z":0.2},{"d":0.5,"facing":90,"h":0.85,"id":"c4-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":9.75,"z":2.37},{"d":0.35,"h":1.1,"id":"c4-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":11.099999999999998,"z":0.2},{"d":0.65,"h":0.85,"id":"treat-trolley","kind":"platform_trolley","label":"Clinical trolley","note":"Platform trolley proxy; client to confirm clinical trolley model and parking position.","source":"assumed","w":0.55,"x":0.2,"z":8.05},{"d":0.4,"h":0.65,"id":"treat-cabinet-2","kind":"cabinet","label":"Treatment cabinet","note":"Client to confirm wall cabinet storage and mounting height.","source":"assumed","w":0.9,"x":1.3,"y":1.5,"z":7.05},{"color":12245203,"d":1.8,"h":1.95,"id":"treat-curtain","kind":"screen","label":"Privacy curtain between couches","note":"Screen proxy for retractable curtain; client to confirm privacy coverage and track arrangement.","source":"assumed","w":0.06,"x":2.15,"z":9.15},{"color":15059361,"d":1.3,"h":0.01,"id":"kids-rug","kind":"rug","label":"Kids corner rug","note":"Client to confirm washable, slip-resistant rug and cleaning policy.","source":"assumed","w":2.2,"x":17.6,"y":0.025,"z":12.75},{"d":0.4,"h":1.1,"id":"wait-plant-1","kind":"plant","label":"Waiting plant","note":"Client to confirm planting and infection-control/maintenance policy.","source":"assumed","w":0.4,"x":0.2,"z":11.95},{"d":0.4,"h":1.1,"id":"wait-plant-2","kind":"plant","label":"Waiting plant","note":"Client to confirm planting and maintenance; keep approach clear.","source":"assumed","w":0.4,"x":7.75,"z":11.3},{"d":0.4,"h":1.1,"id":"wait-water","kind":"machine","label":"Water dispenser","note":"Machine proxy; client to confirm dispenser, supply, drip tray and servicing.","source":"assumed","w":0.45,"x":7.75,"z":13.65},{"d":0.6,"h":1,"id":"reception-files","kind":"cabinet","label":"Lockable files cabinet","note":"Client to confirm records capacity, locks and staff access.","source":"assumed","w":0.4,"x":12.85,"z":11.85},{"d":0.3,"h":0.4,"id":"reception-workstation-1","kind":"pos_terminal","label":"Workstation 1 · check-in + payment","note":"Terminal proxy for computer; client to confirm IT, cabling and both staff handling check-in/payment.","source":"assumed","w":0.4,"x":13.35,"y":1.05,"z":12.65},{"d":0.3,"h":0.4,"id":"reception-workstation-2","kind":"pos_terminal","label":"Workstation 2 · check-in + payment","note":"Terminal proxy for computer; client to confirm IT, cabling and both staff handling check-in/payment.","source":"assumed","w":0.4,"x":14.85,"y":1.05,"z":12.65},{"d":0.25,"h":0.2,"id":"reception-payment","kind":"pos_terminal","label":"Shared payment terminal","note":"Client to confirm terminal reach, sharing by both receptionists and accessible counter knee recess.","source":"assumed","w":0.25,"x":15.8,"y":0.76,"z":12.9}],"radius":0.24,"scaleConfirmed":false,"serviceSeconds":15,"speed":0.6,"task":"clean","width":21},"x":0,"z":0}],"links":[],"name":"Linden Street Clinic / UC-MED-01 A — shared reception","storyHeight":4},"capturedAt":"2026-10-09T12:14:53.805Z","comparisons":[{"baseline":{"building":{"activeFloor":"floor-1","floors":[{"id":"floor-1","level":0,"scene":{"assumptions":["Agent-created spatial draft. Verify object footprints and passage widths.","Robot settings, workload and station positions are illustrative defaults."],"coverageWidth":0.6,"demandPerHour":30,"depth":15,"destination":{"x":10.5,"z":6},"dock":{"x":10.5,"z":14.1},"durationMinutes":30,"fleetSize":1,"kind":"facility","name":"Linden Street Clinic / UC-MED-01 A — shared reception","objects":[{"color":14410462,"d":0.2,"h":3,"id":"north","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":21,"x":0,"z":0},{"color":14410462,"d":14.6,"h":3,"id":"west","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.2,"x":0,"z":0.2},{"color":14410462,"d":14.6,"h":3,"id":"east","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.2,"x":20.8,"z":0.2},{"color":14410462,"d":0.2,"h":3,"id":"south","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":21,"x":0,"z":14.8},{"color":11919333,"d":0.2,"h":2.9,"id":"glaze-w","kind":"window","label":"Street glazing","note":"Full-height shopfront assumed; mullions to survey.","source":"assumed","w":9.4,"x":0.2,"y":0.05,"z":14.8},{"color":11919333,"d":0.2,"h":2.9,"id":"glaze-e","kind":"window","label":"Street glazing","note":"Full-height shopfront assumed; mullions to survey.","source":"assumed","w":9.4,"x":11.4,"y":0.05,"z":14.8},{"d":0.2,"doorType":"automatic","h":2.2,"id":"entrance","kind":"door","label":"Entrance 1.8 m automatic","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.8,"x":9.6,"z":14.8},{"d":0.2,"doorType":"hinged","h":2.2,"id":"service","kind":"door","label":"Staff / service exit 1.1 m","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.1,"x":19.1,"z":0},{"color":6911869,"d":0.5,"h":3,"id":"col-7-5","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":6.75,"z":4.75},{"color":6911869,"d":0.5,"h":3,"id":"col-7-10","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":6.75,"z":9.75},{"color":6911869,"d":0.5,"h":3,"id":"col-14-5","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":13.75,"z":4.75},{"color":6911869,"d":0.5,"h":3,"id":"col-14-10","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":13.75,"z":9.75},{"color":14410462,"d":3.6,"h":3,"id":"c1-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":3.2,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c1-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":0.2,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c1","kind":"zone","label":"Consult 1 · 3.0 × 3.6","source":"assumed","w":3,"x":0.2,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c1-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":0.35,"z":3.8},{"d":0.6,"h":0.75,"id":"c1-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":0.4,"z":0.95},{"d":0.5,"h":0.85,"id":"c1-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":0.65,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c1-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":0.65,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c1-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":2.5,"z":0.2},{"d":0.45,"h":0.85,"id":"c1-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":2.6,"z":3.15},{"color":14410462,"d":3.6,"h":3,"id":"c2-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":6.35,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c2-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":3.35,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c2","kind":"zone","label":"Consult 2 · 3.0 × 3.6","source":"assumed","w":3,"x":3.35,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c2-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":3.5,"z":3.8},{"d":0.6,"h":0.75,"id":"c2-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":3.5500000000000003,"z":0.95},{"d":0.5,"h":0.85,"id":"c2-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":3.8000000000000003,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c2-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":3.8,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c2-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":5.6499999999999995,"z":0.2},{"d":0.45,"h":0.85,"id":"c2-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":5.75,"z":3.15},{"color":14410462,"d":3.6,"h":3,"id":"c3-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":9.5,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c3-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":6.5,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c3","kind":"zone","label":"Consult 3 · 3.0 × 3.6","source":"assumed","w":3,"x":6.5,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c3-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":7.55,"z":3.8},{"d":0.6,"h":0.75,"id":"c3-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":6.7,"z":0.95},{"d":0.5,"h":0.85,"id":"c3-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":6.95,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c3-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":6.95,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c3-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":8.8,"z":0.2},{"d":0.45,"h":0.85,"id":"c3-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":8.899999999999999,"z":3.15},{"color":14410462,"d":3.6,"h":3,"id":"c4-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":12.649999999999999,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c4-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":9.649999999999999,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c4","kind":"zone","label":"Consult 4 · 3.0 × 3.6","source":"assumed","w":3,"x":9.649999999999999,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c4-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":9.799999999999999,"z":3.8},{"d":0.6,"h":0.75,"id":"c4-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":9.849999999999998,"z":0.95},{"d":0.5,"h":0.85,"id":"c4-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":10.099999999999998,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c4-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":10.1,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c4-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":11.95,"z":0.2},{"d":0.45,"h":0.85,"id":"c4-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":12.049999999999999,"z":3.15},{"color":14410462,"d":4.5,"h":3,"id":"staff-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":14.5,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"staff-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":6.3,"x":14.5,"z":4.7},{"color":15196113,"d":4.5,"h":0.02,"id":"staff","kind":"zone","label":"Staff room / lockers / kitchenette","source":"assumed","w":6.15,"x":14.65,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"staff-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":15,"z":4.7},{"d":0.5,"h":1.8,"id":"locker-0","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":14.85,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-1","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":15.33,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-2","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":15.809999999999999,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-3","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":16.29,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-4","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":16.77,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-5","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":17.25,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-6","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":17.73,"z":0.4},{"d":0.6,"h":0.9,"id":"kitchen","kind":"kitchen","label":"Kitchenette","note":"Client to confirm kitchenette dimensions, specification and placement.","source":"assumed","w":2.4,"x":18.15,"z":4.1},{"d":0.65,"h":1.8,"id":"fridge","kind":"fridge","label":"Fridge","note":"Client to confirm fridge dimensions, specification and placement.","source":"assumed","w":0.65,"x":20.15,"z":3.05},{"d":0.75,"h":0.74,"id":"staff-table","kind":"table","label":"Staff table","note":"Client to confirm staff table dimensions, specification and placement.","source":"assumed","w":1.5,"x":16.65,"z":2.05},{"d":0.5,"h":0.85,"id":"staff-chair-0-0","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":16.65,"z":1.35},{"d":0.5,"facing":180,"h":0.85,"id":"staff-chair-0-1","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":16.65,"z":3.15},{"d":0.5,"h":0.85,"id":"staff-chair-1-0","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":17.5,"z":1.35},{"d":0.5,"facing":180,"h":0.85,"id":"staff-chair-1-1","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":17.5,"z":3.15},{"color":15789020,"d":1.65,"h":0.02,"id":"clinical","kind":"zone","label":"Clinical circulation · 1.65 m min main lane","source":"assumed","tags":["passage"],"w":16.45,"x":0.2,"z":5.25},{"color":15789020,"d":4.8,"h":0.02,"id":"spine","kind":"zone","label":"Patient circulation","source":"assumed","tags":["passage"],"w":1.85,"x":9.45,"z":7.05},{"color":14410462,"d":4,"h":3,"id":"treat-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":4.2,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"treat-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":0.05000000000000002,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"treat-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":0.05000000000000002,"z":11.05},{"color":14018790,"d":4,"h":0.02,"id":"treat","kind":"zone","label":"Treatment · 4.0 × 4.0","source":"assumed","w":4,"x":0.2,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"treat-door","kind":"door","label":"Treatment 1.2 m clear","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.2,"x":2.85,"z":6.9},{"color":12703967,"d":2,"h":0.65,"id":"treat-bed-0","kind":"exam_couch","label":"Treatment couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.75,"x":0.2,"z":9.05},{"color":12703967,"d":2,"h":0.65,"id":"treat-bed-1","kind":"exam_couch","label":"Treatment couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.75,"x":3.45,"z":9.05},{"blocks":false,"d":1.8,"h":0.3,"id":"treat-screen","kind":"partition","label":"Ceiling privacy curtain track","note":"Client to confirm ceiling privacy curtain track dimensions, specification and placement.","source":"assumed","w":0.08,"x":2.15,"y":2.2,"z":9.15},{"d":0.5,"h":0.85,"id":"treat-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.6,"x":0.5,"z":7.3},{"d":0.45,"h":1,"id":"treat-supply","kind":"cabinet","label":"Clinical supplies","note":"Client to confirm clinical supplies dimensions, specification and placement.","source":"assumed","w":1,"x":1.25,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"awc-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":4.35,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"awc-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":6.9,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"awc-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.6999999999999997,"x":4.35,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"awc-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.6999999999999997,"x":4.35,"z":9.45},{"color":14736624,"d":2.4,"h":0.02,"id":"awc","kind":"zone","label":"Accessible WC · 2.4 × 2.4","source":"assumed","w":2.4,"x":4.5,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"awc-door","kind":"door","label":"Accessible WC 1.0 m clear","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":5.5,"z":6.9},{"d":0.7,"facing":180,"h":0.8,"id":"awc-toilet","kind":"toilet","label":"WC with side-transfer space","note":"Client to confirm grab rails: locations, heights, fixings and transfer side with accessibility specialist; retain clear turning area.","source":"assumed","w":0.4,"x":4.75,"z":8.55},{"d":0.45,"h":0.8,"id":"awc-basin","kind":"washbasin","label":"Accessible basin","note":"Client to confirm accessible basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":4.65,"z":7.3},{"color":15854840,"d":1.5,"footprint":[{"x":1.5,"z":0.75},{"x":1.4855889603024228,"z":0.8963177415120962},{"x":1.442909649383465,"z":1.0370125742738172},{"x":1.373602209226909,"z":1.1666776747647016},{"x":1.2803300858899107,"z":1.2803300858899105},{"x":1.1666776747647019,"z":1.373602209226909},{"x":1.0370125742738174,"z":1.442909649383465},{"x":0.8963177415120962,"z":1.4855889603024228},{"x":0.75,"z":1.5},{"x":0.6036822584879038,"z":1.4855889603024228},{"x":0.4629874257261827,"z":1.442909649383465},{"x":0.33332232523529853,"z":1.3736022092269091},{"x":0.2196699141100894,"z":1.2803300858899107},{"x":0.126397790773091,"z":1.1666776747647016},{"x":0.057090350616535,"z":1.0370125742738174},{"x":0.014411039697577177,"z":0.8963177415120964},{"x":0,"z":0.7500000000000001},{"x":0.014411039697577177,"z":0.6036822584879038},{"x":0.05709035061653489,"z":0.46298742572618273},{"x":0.12639779077309088,"z":0.33332232523529853},{"x":0.2196699141100893,"z":0.2196699141100894},{"x":0.33332232523529837,"z":0.1263977907730911},{"x":0.46298742572618223,"z":0.05709035061653511},{"x":0.6036822584879035,"z":0.014411039697577288},{"x":0.7499999999999999,"z":0},{"x":0.8963177415120962,"z":0.014411039697577177},{"x":1.0370125742738174,"z":0.057090350616535},{"x":1.1666776747647014,"z":0.12639779077309088},{"x":1.2803300858899105,"z":0.2196699141100893},{"x":1.373602209226909,"z":0.33332232523529837},{"x":1.442909649383465,"z":0.46298742572618223},{"x":1.4855889603024228,"z":0.6036822584879035}],"h":0.02,"id":"awc-turn","kind":"zone","label":"Ø 1.5 m clear turning circle","source":"assumed","symbol":"wheelchair","w":1.5,"x":5.25,"z":7.65},{"color":14410462,"d":2.4,"h":3,"id":"wc-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":7.35,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"wc-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":9.3,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"wc-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1,"x":7.35,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"wc-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1,"x":7.35,"z":9.45},{"color":14736624,"d":2.4,"h":0.02,"id":"wc","kind":"zone","label":"WC · 1.8 × 2.4","source":"assumed","w":1.8,"x":7.5,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"wc-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":7.65,"z":6.9},{"d":0.7,"facing":180,"h":0.8,"id":"wc-toilet","kind":"toilet","label":"WC","note":"Client to confirm wc dimensions, specification and placement.","source":"assumed","w":0.4,"x":8.55,"z":8.55},{"d":0.4,"h":0.85,"id":"wc-basin","kind":"washbasin","label":"Basin","note":"Client to confirm basin dimensions, specification and placement.","source":"assumed","w":0.45,"x":8.65,"z":7.3},{"color":14410462,"d":2.4,"h":3,"id":"utility-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":11.299999999999999,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"utility-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":14.45,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"utility-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3.3,"x":11.299999999999999,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"utility-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3.3,"x":11.299999999999999,"z":9.45},{"color":14871000,"d":2.4,"h":0.02,"id":"utility","kind":"zone","label":"Nurse / clean utility","source":"assumed","w":3,"x":11.45,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"utility-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":11.8,"z":6.9},{"d":0.55,"h":0.9,"id":"utility-work","kind":"counter","label":"Nurse worktop","note":"Client to confirm nurse worktop dimensions, specification and placement.","source":"assumed","w":2.45,"x":11.7,"z":8.9},{"d":0.45,"h":0.85,"id":"utility-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.55,"x":13.65,"z":7.3},{"color":14410462,"d":2.2,"h":3,"id":"control-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":14.6,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"control-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1999999999999997,"x":14.6,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"control-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1999999999999997,"x":14.6,"z":9.25},{"color":14475241,"d":2.2,"h":0.02,"id":"control","kind":"zone","label":"X-ray control","source":"assumed","w":1.9,"x":14.75,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"control-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":14.9,"z":6.9},{"d":1.3,"h":0.75,"id":"control-console","kind":"desk","label":"Control console","note":"Client to confirm control console dimensions, specification and placement.","source":"assumed","w":0.55,"x":16.1,"z":7.35},{"d":0.5,"facing":90,"h":0.85,"id":"control-chair","kind":"chair","label":"Radiographer","note":"Client to confirm radiographer dimensions, specification and placement.","source":"assumed","w":0.5,"x":15.1,"z":8.35},{"color":14410462,"d":4.5,"h":3,"id":"xray-w","kind":"wall","label":"Control shielding screen · shielding by specialist","note":"shielding by specialist; confirm barrier, glazing, door and equipment operating envelopes.","source":"assumed","w":0.15,"x":16.650000000000002,"z":6.55},{"color":14410462,"d":0.15,"h":3,"id":"xray-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":16.650000000000002,"z":6.3999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"xray-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":16.650000000000002,"z":11.049999999999999},{"color":14475241,"d":4.5,"h":0.02,"id":"xray","kind":"zone","label":"X-ray · 4.0 × 4.5","source":"assumed","w":4,"x":16.8,"z":6.55},{"d":1.2,"doorType":"sliding","h":2.2,"id":"xray-door","kind":"door","label":"X-ray 1.2 m clear","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":0.15,"x":16.65,"z":9.75},{"d":1.1,"h":0.8,"id":"control-window","kind":"window","label":"Specialist shielding / viewing glass","note":"All X-ray shielding, door, glazing and equipment clearances by radiation specialist.","source":"assumed","w":0.15,"x":16.65,"y":1.1,"z":7.45},{"d":1.2,"h":2,"id":"xray-table","kind":"xray_unit","label":"X-ray unit / representative equipment","note":"Current X-ray unit replaces the old table and box stand-ins. Assumed supplier envelope; radiation shielding, clinical use and operating clearances are not validated.","source":"assumed","w":2.4,"x":18.2,"z":8.1},{"d":0.45,"h":1.1,"id":"xray-store","kind":"cabinet","label":"Imaging storage","note":"Client to confirm imaging storage dimensions, specification and placement.","source":"assumed","w":1.1,"x":17.15,"z":6.55},{"color":15067865,"d":3.55,"h":0.02,"id":"waiting","kind":"zone","label":"Waiting · 20 seats + 2 wheelchair spaces","source":"assumed","w":8.05,"x":0.2,"z":11.25},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-0","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":0.5,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-1","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":1.1800000000000002,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-2","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":1.86,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-3","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":2.54,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-4","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.22,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-5","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.9000000000000004,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-6","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":4.58,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-7","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.260000000000001,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-8","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.94,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-9","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":6.62,"z":11.3},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-0","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":2.65,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-1","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.2,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-2","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.75,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-3","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":4.3,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-4","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":4.85,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-5","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.4,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-6","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.95,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-7","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":6.5,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-8","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":7.05,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-9","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":7.6,"z":14.15},{"color":13098711,"d":1.4,"h":0.02,"id":"wheel-0","kind":"zone","label":"Wheelchair","note":"Keep clear. Client to confirm wheelchair bay dimensions and approach.","source":"assumed","symbol":"wheelchair","w":0.9,"x":0.5,"z":13.35},{"color":13098711,"d":1.4,"h":0.02,"id":"wheel-1","kind":"zone","label":"Wheelchair","note":"Keep clear. Client to confirm wheelchair bay dimensions and approach.","source":"assumed","symbol":"wheelchair","w":0.9,"x":1.65,"z":13.35},{"color":14410462,"d":2.8,"h":3,"id":"lobby-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":8.25,"z":12},{"color":14410462,"d":2.8,"h":3,"id":"lobby-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":12.6,"z":12},{"color":14410462,"d":0.15,"h":3,"id":"lobby-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.5,"x":8.25,"z":11.85},{"color":15002091,"d":2.8,"h":0.02,"id":"lobby","kind":"zone","label":"Wind lobby · wheelchair / buggy parking","source":"assumed","w":4.2,"x":8.4,"z":12},{"d":0.15,"doorType":"automatic","h":2.2,"id":"lobby-inner","kind":"door","label":"Lobby 1.8 m automatic","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.8,"x":9.6,"z":11.85},{"color":13098711,"d":1.4,"h":0.02,"id":"lobby-wc","kind":"zone","label":"Wheelchair park","source":"assumed","symbol":"wheelchair","w":0.85,"x":8.55,"z":13.35},{"color":14278618,"d":1.3,"h":0.02,"id":"lobby-buggy","kind":"zone","label":"Buggy park","source":"assumed","w":0.8,"x":11.6,"z":13.35},{"color":15261127,"d":2.1,"h":0.02,"id":"reception","kind":"zone","label":"Reception / payment · 2 staff","source":"assumed","w":3.5,"x":12.9,"z":11.4},{"color":11836537,"d":0.7,"h":1.05,"id":"reception-counter","kind":"counter","label":"Reception / payment","note":"Client to confirm reception / payment dimensions, specification and placement.","source":"assumed","w":2.6,"x":12.75,"z":12.6},{"d":0.7,"h":0.76,"id":"reception-low","kind":"counter","label":"Accessible counter","note":"Lower section; detail knee recess and hearing loop.","source":"assumed","w":1,"x":15.35,"z":12.6},{"d":0.55,"h":0.85,"id":"recept-chair-0","kind":"chair","label":"Receptionist","note":"Client to confirm receptionist dimensions, specification and placement.","source":"assumed","w":0.55,"x":13.35,"z":11.8},{"d":0.55,"h":0.85,"id":"recept-chair-1","kind":"chair","label":"Receptionist","note":"Client to confirm receptionist dimensions, specification and placement.","source":"assumed","w":0.55,"x":14.95,"z":11.8},{"color":15851979,"d":2.5,"h":0.02,"id":"kids","kind":"zone","label":"Kids corner","source":"assumed","w":3.3,"x":17.2,"z":12.1},{"d":0.7,"h":0.5,"id":"kids-table","kind":"round_table","label":"Kids table","note":"Client to confirm kids table dimensions, specification and placement.","source":"assumed","w":0.7,"x":18.4,"z":13},{"d":0.35,"h":0.3,"id":"kids-stool-0","kind":"stool","label":"Kids seat","note":"Client to confirm kids seat dimensions, specification and placement.","source":"assumed","w":0.35,"x":17.9,"z":13.2},{"d":0.35,"h":0.3,"id":"kids-stool-1","kind":"stool","label":"Kids seat","note":"Client to confirm kids seat dimensions, specification and placement.","source":"assumed","w":0.35,"x":19.2,"z":13.2},{"d":1.5,"h":0.8,"id":"kids-shelf","kind":"bookshelf","label":"Books / toys","note":"Client to confirm books / toys dimensions, specification and placement.","source":"assumed","w":0.35,"x":20.45,"z":12.4},{"color":15196113,"d":3.6,"h":0.02,"id":"linen","kind":"zone","label":"Linen / storage · 1.7 × 3.6","source":"assumed","w":1.7,"x":12.8,"z":0.2},{"d":0.15,"h":3,"id":"linen-s","kind":"wall","label":"Storage partition","source":"assumed","w":1.7,"x":12.8,"z":3.8},{"d":0.15,"doorType":"sliding","h":2.2,"id":"linen-door","kind":"door","label":"Storage 1.0 m clear","source":"assumed","w":1,"x":13,"z":3.8},{"d":2.2,"h":1.8,"id":"linen-shelves","kind":"shelf","label":"Linen shelving","note":"Client to confirm linen shelving dimensions, specification and placement.","source":"assumed","w":0.5,"x":14,"z":0.2},{"color":14410462,"d":2.4,"h":3,"id":"toilet-chase","kind":"block","label":"Plumbing service void","source":"assumed","w":0.3,"x":7.05,"z":7.05},{"d":1.55,"h":2.75,"id":"lobby-glass-w","kind":"window","label":"Glazed lobby screen","source":"assumed","w":0.15,"x":8.25,"y":0.1,"z":13.2},{"d":2.7,"h":2.75,"id":"lobby-glass-e","kind":"window","label":"Glazed lobby screen","source":"assumed","w":0.15,"x":12.6,"y":0.1,"z":12.05},{"d":0.15,"h":2.75,"id":"lobby-glass-nw","kind":"window","label":"Lobby glazing","source":"assumed","w":1.2,"x":8.4,"y":0.1,"z":11.85},{"d":0.15,"h":2.75,"id":"lobby-glass-ne","kind":"window","label":"Lobby glazing","source":"assumed","w":1.2,"x":11.4,"y":0.1,"z":11.85},{"d":1.1,"doorType":"sliding","h":2.2,"id":"waiting-door","kind":"door","label":"Waiting 1.1 m clear sliding","source":"assumed","w":0.15,"x":8.25,"z":12.1},{"d":1.5,"h":0.02,"id":"xray-approach","kind":"zone","label":"X-ray approach · 1.5 m clear","source":"assumed","tags":["passage"],"w":5.35,"x":11.3,"z":10.25},{"d":0.5,"facing":90,"h":0.85,"id":"c1-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":0.3,"z":2.37},{"d":0.35,"h":1.1,"id":"c1-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":1.65,"z":0.2},{"d":0.5,"facing":90,"h":0.85,"id":"c2-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":3.45,"z":2.37},{"d":0.35,"h":1.1,"id":"c2-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":4.8,"z":0.2},{"d":0.5,"facing":90,"h":0.85,"id":"c3-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":6.6,"z":2.37},{"d":0.35,"h":1.1,"id":"c3-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":7.95,"z":0.2},{"d":0.5,"facing":90,"h":0.85,"id":"c4-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":9.75,"z":2.37},{"d":0.35,"h":1.1,"id":"c4-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":11.099999999999998,"z":0.2},{"d":0.65,"h":0.85,"id":"treat-trolley","kind":"platform_trolley","label":"Clinical trolley","note":"Platform trolley proxy; client to confirm clinical trolley model and parking position.","source":"assumed","w":0.55,"x":0.2,"z":8.05},{"d":0.4,"h":0.65,"id":"treat-cabinet-2","kind":"cabinet","label":"Treatment cabinet","note":"Client to confirm wall cabinet storage and mounting height.","source":"assumed","w":0.9,"x":1.3,"y":1.5,"z":7.05},{"color":12245203,"d":1.8,"h":1.95,"id":"treat-curtain","kind":"screen","label":"Privacy curtain between couches","note":"Screen proxy for retractable curtain; client to confirm privacy coverage and track arrangement.","source":"assumed","w":0.06,"x":2.15,"z":9.15},{"color":15059361,"d":1.3,"h":0.01,"id":"kids-rug","kind":"rug","label":"Kids corner rug","note":"Client to confirm washable, slip-resistant rug and cleaning policy.","source":"assumed","w":2.2,"x":17.6,"y":0.025,"z":12.75},{"d":0.4,"h":1.1,"id":"wait-plant-1","kind":"plant","label":"Waiting plant","note":"Client to confirm planting and infection-control/maintenance policy.","source":"assumed","w":0.4,"x":0.2,"z":11.95},{"d":0.4,"h":1.1,"id":"wait-plant-2","kind":"plant","label":"Waiting plant","note":"Client to confirm planting and maintenance; keep approach clear.","source":"assumed","w":0.4,"x":7.75,"z":11.3},{"d":0.4,"h":1.1,"id":"wait-water","kind":"machine","label":"Water dispenser","note":"Machine proxy; client to confirm dispenser, supply, drip tray and servicing.","source":"assumed","w":0.45,"x":7.75,"z":13.65},{"d":0.6,"h":1,"id":"reception-files","kind":"cabinet","label":"Lockable files cabinet","note":"Client to confirm records capacity, locks and staff access.","source":"assumed","w":0.4,"x":12.85,"z":11.85},{"d":0.3,"h":0.4,"id":"reception-workstation-1","kind":"pos_terminal","label":"Workstation 1 · check-in + payment","note":"Terminal proxy for computer; client to confirm IT, cabling and both staff handling check-in/payment.","source":"assumed","w":0.4,"x":13.35,"y":1.05,"z":12.65},{"d":0.3,"h":0.4,"id":"reception-workstation-2","kind":"pos_terminal","label":"Workstation 2 · check-in + payment","note":"Terminal proxy for computer; client to confirm IT, cabling and both staff handling check-in/payment.","source":"assumed","w":0.4,"x":14.85,"y":1.05,"z":12.65},{"d":0.25,"h":0.2,"id":"reception-payment","kind":"pos_terminal","label":"Shared payment terminal","note":"Client to confirm terminal reach, sharing by both receptionists and accessible counter knee recess.","source":"assumed","w":0.25,"x":15.8,"y":0.76,"z":12.9}],"radius":0.24,"scaleConfirmed":false,"serviceSeconds":15,"speed":0.6,"task":"clean","width":21},"x":0,"z":0}],"links":[],"name":"Linden Street Clinic / UC-MED-01 A — shared reception","storyHeight":4},"createdAt":"2026-10-09T12:03:07.611Z","id":"f14240ba-9d6a-418a-a707-53a455617d01","name":"UC-MED-01 A / 4 rooms / 7 staff / 20 paired seeds","scenario":{"agents":[{"appearance":"person","height":1.7,"id":"patient","radius":0.3,"speed":1}],"arrivals":[{"at":{"floor":"floor-1","point":{"x":10.5,"z":14.1}},"id":"morning-arrivals","mode":"poisson","processId":"morning","profile":[{"durationSeconds":3600,"ratePerHour":24},{"durationSeconds":3600,"ratePerHour":16},{"durationSeconds":1800,"ratePerHour":10}],"startSeconds":0}],"horizonSeconds":21600,"id":"UC-MED-A-morning","maxEvents":100000,"name":"A / current four-room clinic / 20 paired seeds","parameters":{},"processes":[{"agent":"patient","id":"morning","steps":[{"to":{"floor":"floor-1","object":"reception-counter","side":"south"},"type":"goto"},{"count":1,"processId":"checkin-wait","type":"call"},{"durationSeconds":{"max":180,"min":60,"random":"uniform"},"type":"delay"},{"resourceId":"receptionists","type":"release","units":1},{"to":{"floor":"floor-1","point":{"x":4.225,"z":13.025}},"type":"goto"},{"count":1,"processId":"doctor-wait","type":"call"},{"count":1,"processId":"consult-room","type":"call"},{"resourceId":"doctors","type":"release","units":1},{"count":1,"processId":"after-consult","type":"call"},{"to":{"floor":"floor-1","object":"reception-counter","side":"south"},"type":"goto"},{"count":1,"processId":"payment-wait","type":"call"},{"durationSeconds":{"max":120,"min":60,"random":"uniform"},"type":"delay"},{"resourceId":"receptionists","type":"release","units":1},{"to":{"floor":"floor-1","point":{"x":10.5,"z":14.1}},"type":"goto"},{"type":"finish"}]},{"id":"checkin-wait","steps":[{"resourceId":"receptionists","type":"acquire","units":1},{"type":"finish"}]},{"id":"payment-wait","steps":[{"resourceId":"receptionists","type":"acquire","units":1},{"type":"finish"}]},{"id":"doctor-wait","steps":[{"resourceId":"doctors","type":"acquire","units":1},{"type":"finish"}]},{"id":"return-doctor-wait","steps":[{"resourceId":"doctors","type":"acquire","units":1},{"type":"finish"}]},{"id":"consult-room","steps":[{"by":"free","options":[{"processId":"consult1","resourceId":"room1"},{"processId":"consult2","resourceId":"room2"},{"processId":"consult3","resourceId":"room3"},{"processId":"consult4","resourceId":"room4"}],"type":"branch"}]},{"id":"review-room","steps":[{"by":"free","options":[{"processId":"review1","resourceId":"room1"},{"processId":"review2","resourceId":"room2"},{"processId":"review3","resourceId":"room3"},{"processId":"review4","resourceId":"room4"}],"type":"branch"}]},{"id":"after-consult","steps":[{"options":[{"processId":"none","weight":70},{"processId":"treatment","weight":20},{"processId":"imaging","weight":10}],"type":"branch"}]},{"id":"none","steps":[{"type":"finish"}]},{"id":"treatment","steps":[{"to":{"floor":"floor-1","point":{"x":4.225,"z":13.025}},"type":"goto"},{"resourceId":"nurses","type":"acquire","units":1},{"to":{"floor":"floor-1","point":{"x":2.2,"z":8.5}},"type":"goto"},{"durationSeconds":{"max":900,"min":300,"random":"uniform"},"type":"delay"},{"resourceId":"nurses","type":"release","units":1},{"type":"finish"}]},{"id":"imaging","steps":[{"to":{"floor":"floor-1","point":{"x":4.225,"z":13.025}},"type":"goto"},{"resourceId":"radiographer","type":"acquire","units":1},{"to":{"floor":"floor-1","point":{"x":19.4,"z":9.9}},"type":"goto"},{"durationSeconds":{"max":600,"min":300,"random":"uniform"},"type":"delay"},{"resourceId":"radiographer","type":"release","units":1},{"to":{"floor":"floor-1","point":{"x":4.225,"z":13.025}},"type":"goto"},{"count":1,"processId":"return-doctor-wait","type":"call"},{"count":1,"processId":"review-room","type":"call"},{"resourceId":"doctors","type":"release","units":1},{"type":"finish"}]},{"id":"consult1","steps":[{"resourceId":"room1","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c1"},"type":"goto"},{"durationSeconds":{"max":720,"min":240,"mode":420,"random":"triangular"},"type":"delay"},{"resourceId":"room1","type":"release","units":1},{"type":"finish"}]},{"id":"review1","steps":[{"resourceId":"room1","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c1"},"type":"goto"},{"durationSeconds":{"max":240,"min":120,"random":"uniform"},"type":"delay"},{"resourceId":"room1","type":"release","units":1},{"type":"finish"}]},{"id":"consult2","steps":[{"resourceId":"room2","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c2"},"type":"goto"},{"durationSeconds":{"max":720,"min":240,"mode":420,"random":"triangular"},"type":"delay"},{"resourceId":"room2","type":"release","units":1},{"type":"finish"}]},{"id":"review2","steps":[{"resourceId":"room2","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c2"},"type":"goto"},{"durationSeconds":{"max":240,"min":120,"random":"uniform"},"type":"delay"},{"resourceId":"room2","type":"release","units":1},{"type":"finish"}]},{"id":"consult3","steps":[{"resourceId":"room3","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c3"},"type":"goto"},{"durationSeconds":{"max":720,"min":240,"mode":420,"random":"triangular"},"type":"delay"},{"resourceId":"room3","type":"release","units":1},{"type":"finish"}]},{"id":"review3","steps":[{"resourceId":"room3","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c3"},"type":"goto"},{"durationSeconds":{"max":240,"min":120,"random":"uniform"},"type":"delay"},{"resourceId":"room3","type":"release","units":1},{"type":"finish"}]},{"id":"consult4","steps":[{"resourceId":"room4","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c4"},"type":"goto"},{"durationSeconds":{"max":720,"min":240,"mode":420,"random":"triangular"},"type":"delay"},{"resourceId":"room4","type":"release","units":1},{"type":"finish"}]},{"id":"review4","steps":[{"resourceId":"room4","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c4"},"type":"goto"},{"durationSeconds":{"max":240,"min":120,"random":"uniform"},"type":"delay"},{"resourceId":"room4","type":"release","units":1},{"type":"finish"}]}],"resources":[{"capacity":2,"id":"receptionists","name":"Two shared reception/payment staff","role":"shared"},{"capacity":2,"id":"doctors","role":"shared"},{"capacity":2,"id":"nurses","role":"shared"},{"capacity":1,"id":"radiographer","role":"shared"},{"capacity":1,"id":"room1","role":"equipment"},{"capacity":1,"id":"room2","role":"equipment"},{"capacity":1,"id":"room3","role":"equipment"},{"capacity":1,"id":"room4","role":"equipment"},{"at":{"floor":"floor-1","object":"waiting"},"capacity":1000,"id":"waiting-demand","mode":"occupy","role":"zone"}],"runs":20,"seed":20261009,"traceLimit":10000,"version":5,"warmupSeconds":0},"scenarioHash":"fnv1a-56733229","spaceHash":"fnv1a-11feab13"},"candidate":{"building":{"activeFloor":"floor-1","floors":[{"id":"floor-1","level":0,"scene":{"assumptions":["Agent-created spatial draft. Verify object footprints and passage widths.","Robot settings, workload and station positions are illustrative defaults."],"coverageWidth":0.6,"demandPerHour":30,"depth":15,"destination":{"x":10.5,"z":6},"dock":{"x":10.5,"z":14.1},"durationMinutes":30,"fleetSize":1,"kind":"facility","name":"Linden Street Clinic / UC-MED-01 B — separate payment","objects":[{"color":14410462,"d":0.2,"h":3,"id":"north","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":21,"x":0,"z":0},{"color":14410462,"d":14.6,"h":3,"id":"west","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.2,"x":0,"z":0.2},{"color":14410462,"d":14.6,"h":3,"id":"east","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.2,"x":20.8,"z":0.2},{"color":14410462,"d":0.2,"h":3,"id":"south","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":21,"x":0,"z":14.8},{"color":11919333,"d":0.2,"h":2.9,"id":"glaze-w","kind":"window","label":"Street glazing","note":"Full-height shopfront assumed; mullions to survey.","source":"assumed","w":9.4,"x":0.2,"y":0.05,"z":14.8},{"color":11919333,"d":0.2,"h":2.9,"id":"glaze-e","kind":"window","label":"Street glazing","note":"Full-height shopfront assumed; mullions to survey.","source":"assumed","w":9.4,"x":11.4,"y":0.05,"z":14.8},{"d":0.2,"doorType":"automatic","h":2.2,"id":"entrance","kind":"door","label":"Entrance 1.8 m automatic","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.8,"x":9.6,"z":14.8},{"d":0.2,"doorType":"hinged","h":2.2,"id":"service","kind":"door","label":"Staff / service exit 1.1 m","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.1,"x":19.1,"z":0},{"color":6911869,"d":0.5,"h":3,"id":"col-7-5","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":6.75,"z":4.75},{"color":6911869,"d":0.5,"h":3,"id":"col-7-10","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":6.75,"z":9.75},{"color":6911869,"d":0.5,"h":3,"id":"col-14-5","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":13.75,"z":4.75},{"color":6911869,"d":0.5,"h":3,"id":"col-14-10","kind":"column","label":"EXISTING fixed column","note":"Fixed position and 500 × 500 mm size supplied in brief; not surveyed.","source":"assumed","tags":["existing"],"w":0.5,"x":13.75,"z":9.75},{"color":14410462,"d":3.6,"h":3,"id":"c1-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":3.2,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c1-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":0.2,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c1","kind":"zone","label":"Consult 1 · 3.0 × 3.6","source":"assumed","w":3,"x":0.2,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c1-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":0.35,"z":3.8},{"d":0.6,"h":0.75,"id":"c1-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":0.4,"z":0.95},{"d":0.5,"h":0.85,"id":"c1-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":0.65,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c1-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":0.65,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c1-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":2.5,"z":0.2},{"d":0.45,"h":0.85,"id":"c1-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":2.6,"z":3.15},{"color":14410462,"d":3.6,"h":3,"id":"c2-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":6.35,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c2-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":3.35,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c2","kind":"zone","label":"Consult 2 · 3.0 × 3.6","source":"assumed","w":3,"x":3.35,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c2-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":3.5,"z":3.8},{"d":0.6,"h":0.75,"id":"c2-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":3.5500000000000003,"z":0.95},{"d":0.5,"h":0.85,"id":"c2-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":3.8000000000000003,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c2-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":3.8,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c2-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":5.6499999999999995,"z":0.2},{"d":0.45,"h":0.85,"id":"c2-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":5.75,"z":3.15},{"color":14410462,"d":3.6,"h":3,"id":"c3-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":9.5,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c3-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":6.5,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c3","kind":"zone","label":"Consult 3 · 3.0 × 3.6","source":"assumed","w":3,"x":6.5,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c3-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":7.55,"z":3.8},{"d":0.6,"h":0.75,"id":"c3-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":6.7,"z":0.95},{"d":0.5,"h":0.85,"id":"c3-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":6.95,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c3-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":6.95,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c3-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":8.8,"z":0.2},{"d":0.45,"h":0.85,"id":"c3-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":8.899999999999999,"z":3.15},{"color":14410462,"d":3.6,"h":3,"id":"c4-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":12.649999999999999,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"c4-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3,"x":9.649999999999999,"z":3.8},{"color":14413030,"d":3.6,"h":0.02,"id":"c4","kind":"zone","label":"Consult 4 · 3.0 × 3.6","source":"assumed","w":3,"x":9.649999999999999,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"c4-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":9.799999999999999,"z":3.8},{"d":0.6,"h":0.75,"id":"c4-desk","kind":"desk","label":"Consult desk","note":"Client to confirm consult desk dimensions, specification and placement.","source":"assumed","w":1.15,"x":9.849999999999998,"z":0.95},{"d":0.5,"h":0.85,"id":"c4-dr","kind":"chair","label":"Doctor","note":"Client to confirm doctor dimensions, specification and placement.","source":"assumed","w":0.5,"x":10.099999999999998,"z":0.3},{"d":0.5,"facing":180,"h":0.85,"id":"c4-pt","kind":"chair","label":"Patient","note":"Client to confirm patient dimensions, specification and placement.","source":"assumed","w":0.5,"x":10.1,"z":1.85},{"color":14149867,"d":1.9,"h":0.65,"id":"c4-couch","kind":"exam_couch","label":"Examination couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.7,"x":11.95,"z":0.2},{"d":0.45,"h":0.85,"id":"c4-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":12.049999999999999,"z":3.15},{"color":14410462,"d":4.5,"h":3,"id":"staff-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":14.5,"z":0.2},{"color":14410462,"d":0.15,"h":3,"id":"staff-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":6.3,"x":14.5,"z":4.7},{"color":15196113,"d":4.5,"h":0.02,"id":"staff","kind":"zone","label":"Staff room / lockers / kitchenette","source":"assumed","w":6.15,"x":14.65,"z":0.2},{"d":0.15,"doorType":"sliding","h":2.2,"id":"staff-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":15,"z":4.7},{"d":0.5,"h":1.8,"id":"locker-0","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":14.85,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-1","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":15.33,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-2","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":15.809999999999999,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-3","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":16.29,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-4","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":16.77,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-5","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":17.25,"z":0.4},{"d":0.5,"h":1.8,"id":"locker-6","kind":"locker","label":"Staff locker","note":"Client to confirm staff locker dimensions, specification and placement.","source":"assumed","w":0.45,"x":17.73,"z":0.4},{"d":0.6,"h":0.9,"id":"kitchen","kind":"kitchen","label":"Kitchenette","note":"Client to confirm kitchenette dimensions, specification and placement.","source":"assumed","w":2.4,"x":18.15,"z":4.1},{"d":0.65,"h":1.8,"id":"fridge","kind":"fridge","label":"Fridge","note":"Client to confirm fridge dimensions, specification and placement.","source":"assumed","w":0.65,"x":20.15,"z":3.05},{"d":0.75,"h":0.74,"id":"staff-table","kind":"table","label":"Staff table","note":"Client to confirm staff table dimensions, specification and placement.","source":"assumed","w":1.5,"x":16.65,"z":2.05},{"d":0.5,"h":0.85,"id":"staff-chair-0-0","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":16.65,"z":1.35},{"d":0.5,"facing":180,"h":0.85,"id":"staff-chair-0-1","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":16.65,"z":3.15},{"d":0.5,"h":0.85,"id":"staff-chair-1-0","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":17.5,"z":1.35},{"d":0.5,"facing":180,"h":0.85,"id":"staff-chair-1-1","kind":"chair","label":"Staff chair","note":"Client to confirm staff chair dimensions, specification and placement.","source":"assumed","w":0.5,"x":17.5,"z":3.15},{"color":15789020,"d":1.65,"h":0.02,"id":"clinical","kind":"zone","label":"Clinical circulation · 1.65 m min main lane","source":"assumed","tags":["passage"],"w":16.45,"x":0.2,"z":5.25},{"color":15789020,"d":4.8,"h":0.02,"id":"spine","kind":"zone","label":"Patient circulation","source":"assumed","tags":["passage"],"w":1.85,"x":9.45,"z":7.05},{"color":14410462,"d":4,"h":3,"id":"treat-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":4.2,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"treat-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":0.05000000000000002,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"treat-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":0.05000000000000002,"z":11.05},{"color":14018790,"d":4,"h":0.02,"id":"treat","kind":"zone","label":"Treatment · 4.0 × 4.0","source":"assumed","w":4,"x":0.2,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"treat-door","kind":"door","label":"Treatment 1.2 m clear","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.2,"x":2.85,"z":6.9},{"color":12703967,"d":2,"h":0.65,"id":"treat-bed-0","kind":"exam_couch","label":"Treatment couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.75,"x":0.2,"z":9.05},{"color":12703967,"d":2,"h":0.65,"id":"treat-bed-1","kind":"exam_couch","label":"Treatment couch","note":"Current medical part, at the inherited assumed footprint. Clinical access and supplier clearances require review.","source":"assumed","w":0.75,"x":3.45,"z":9.05},{"blocks":false,"d":1.8,"h":0.3,"id":"treat-screen","kind":"partition","label":"Ceiling privacy curtain track","note":"Client to confirm ceiling privacy curtain track dimensions, specification and placement.","source":"assumed","w":0.08,"x":2.15,"y":2.2,"z":9.15},{"d":0.5,"h":0.85,"id":"treat-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.6,"x":0.5,"z":7.3},{"d":0.45,"h":1,"id":"treat-supply","kind":"cabinet","label":"Clinical supplies","note":"Client to confirm clinical supplies dimensions, specification and placement.","source":"assumed","w":1,"x":1.25,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"awc-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":4.35,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"awc-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":6.9,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"awc-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.6999999999999997,"x":4.35,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"awc-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.6999999999999997,"x":4.35,"z":9.45},{"color":14736624,"d":2.4,"h":0.02,"id":"awc","kind":"zone","label":"Accessible WC · 2.4 × 2.4","source":"assumed","w":2.4,"x":4.5,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"awc-door","kind":"door","label":"Accessible WC 1.0 m clear","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":5.5,"z":6.9},{"d":0.7,"facing":180,"h":0.8,"id":"awc-toilet","kind":"toilet","label":"WC with side-transfer space","note":"Client to confirm grab rails: locations, heights, fixings and transfer side with accessibility specialist; retain clear turning area.","source":"assumed","w":0.4,"x":4.75,"z":8.55},{"d":0.45,"h":0.8,"id":"awc-basin","kind":"washbasin","label":"Accessible basin","note":"Client to confirm accessible basin dimensions, specification and placement.","source":"assumed","w":0.5,"x":4.65,"z":7.3},{"color":15854840,"d":1.5,"footprint":[{"x":1.5,"z":0.75},{"x":1.4855889603024228,"z":0.8963177415120962},{"x":1.442909649383465,"z":1.0370125742738172},{"x":1.373602209226909,"z":1.1666776747647016},{"x":1.2803300858899107,"z":1.2803300858899105},{"x":1.1666776747647019,"z":1.373602209226909},{"x":1.0370125742738174,"z":1.442909649383465},{"x":0.8963177415120962,"z":1.4855889603024228},{"x":0.75,"z":1.5},{"x":0.6036822584879038,"z":1.4855889603024228},{"x":0.4629874257261827,"z":1.442909649383465},{"x":0.33332232523529853,"z":1.3736022092269091},{"x":0.2196699141100894,"z":1.2803300858899107},{"x":0.126397790773091,"z":1.1666776747647016},{"x":0.057090350616535,"z":1.0370125742738174},{"x":0.014411039697577177,"z":0.8963177415120964},{"x":0,"z":0.7500000000000001},{"x":0.014411039697577177,"z":0.6036822584879038},{"x":0.05709035061653489,"z":0.46298742572618273},{"x":0.12639779077309088,"z":0.33332232523529853},{"x":0.2196699141100893,"z":0.2196699141100894},{"x":0.33332232523529837,"z":0.1263977907730911},{"x":0.46298742572618223,"z":0.05709035061653511},{"x":0.6036822584879035,"z":0.014411039697577288},{"x":0.7499999999999999,"z":0},{"x":0.8963177415120962,"z":0.014411039697577177},{"x":1.0370125742738174,"z":0.057090350616535},{"x":1.1666776747647014,"z":0.12639779077309088},{"x":1.2803300858899105,"z":0.2196699141100893},{"x":1.373602209226909,"z":0.33332232523529837},{"x":1.442909649383465,"z":0.46298742572618223},{"x":1.4855889603024228,"z":0.6036822584879035}],"h":0.02,"id":"awc-turn","kind":"zone","label":"Ø 1.5 m clear turning circle","source":"assumed","symbol":"wheelchair","w":1.5,"x":5.25,"z":7.65},{"color":14410462,"d":2.4,"h":3,"id":"wc-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":7.35,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"wc-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":9.3,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"wc-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1,"x":7.35,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"wc-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1,"x":7.35,"z":9.45},{"color":14736624,"d":2.4,"h":0.02,"id":"wc","kind":"zone","label":"WC · 1.8 × 2.4","source":"assumed","w":1.8,"x":7.5,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"wc-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":7.65,"z":6.9},{"d":0.7,"facing":180,"h":0.8,"id":"wc-toilet","kind":"toilet","label":"WC","note":"Client to confirm wc dimensions, specification and placement.","source":"assumed","w":0.4,"x":8.55,"z":8.55},{"d":0.4,"h":0.85,"id":"wc-basin","kind":"washbasin","label":"Basin","note":"Client to confirm basin dimensions, specification and placement.","source":"assumed","w":0.45,"x":8.65,"z":7.3},{"color":14410462,"d":2.4,"h":3,"id":"utility-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":11.299999999999999,"z":7.05},{"color":14410462,"d":2.4,"h":3,"id":"utility-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":14.45,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"utility-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3.3,"x":11.299999999999999,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"utility-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":3.3,"x":11.299999999999999,"z":9.45},{"color":14871000,"d":2.4,"h":0.02,"id":"utility","kind":"zone","label":"Nurse / clean utility","source":"assumed","w":3,"x":11.45,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"utility-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":11.8,"z":6.9},{"d":0.55,"h":0.9,"id":"utility-work","kind":"counter","label":"Nurse worktop","note":"Client to confirm nurse worktop dimensions, specification and placement.","source":"assumed","w":2.45,"x":11.7,"z":8.9},{"d":0.45,"h":0.85,"id":"utility-basin","kind":"washbasin","label":"Hand basin","note":"Client to confirm hand basin dimensions, specification and placement.","source":"assumed","w":0.55,"x":13.65,"z":7.3},{"color":14410462,"d":2.2,"h":3,"id":"control-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":14.6,"z":7.05},{"color":14410462,"d":0.15,"h":3,"id":"control-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1999999999999997,"x":14.6,"z":6.8999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"control-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":2.1999999999999997,"x":14.6,"z":9.25},{"color":14475241,"d":2.2,"h":0.02,"id":"control","kind":"zone","label":"X-ray control","source":"assumed","w":1.9,"x":14.75,"z":7.05},{"d":0.15,"doorType":"sliding","h":2.2,"id":"control-door","kind":"door","label":"1.0 m clear door","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1,"x":14.9,"z":6.9},{"d":1.3,"h":0.75,"id":"control-console","kind":"desk","label":"Control console","note":"Client to confirm control console dimensions, specification and placement.","source":"assumed","w":0.55,"x":16.1,"z":7.35},{"d":0.5,"facing":90,"h":0.85,"id":"control-chair","kind":"chair","label":"Radiographer","note":"Client to confirm radiographer dimensions, specification and placement.","source":"assumed","w":0.5,"x":15.1,"z":8.35},{"color":14410462,"d":4.5,"h":3,"id":"xray-w","kind":"wall","label":"Control shielding screen · shielding by specialist","note":"shielding by specialist; confirm barrier, glazing, door and equipment operating envelopes.","source":"assumed","w":0.15,"x":16.650000000000002,"z":6.55},{"color":14410462,"d":0.15,"h":3,"id":"xray-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":16.650000000000002,"z":6.3999999999999995},{"color":14410462,"d":0.15,"h":3,"id":"xray-s","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.3,"x":16.650000000000002,"z":11.049999999999999},{"color":14475241,"d":4.5,"h":0.02,"id":"xray","kind":"zone","label":"X-ray · 4.0 × 4.5","source":"assumed","w":4,"x":16.8,"z":6.55},{"d":1.2,"doorType":"sliding","h":2.2,"id":"xray-door","kind":"door","label":"X-ray 1.2 m clear","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":0.15,"x":16.65,"z":9.75},{"d":1.1,"h":0.8,"id":"control-window","kind":"window","label":"Specialist shielding / viewing glass","note":"All X-ray shielding, door, glazing and equipment clearances by radiation specialist.","source":"assumed","w":0.15,"x":16.65,"y":1.1,"z":7.45},{"d":1.2,"h":2,"id":"xray-table","kind":"xray_unit","label":"X-ray unit / representative equipment","note":"Current X-ray unit replaces the old table and box stand-ins. Assumed supplier envelope; radiation shielding, clinical use and operating clearances are not validated.","source":"assumed","w":2.4,"x":18.2,"z":8.1},{"d":0.45,"h":1.1,"id":"xray-store","kind":"cabinet","label":"Imaging storage","note":"Client to confirm imaging storage dimensions, specification and placement.","source":"assumed","w":1.1,"x":17.15,"z":6.55},{"color":15067865,"d":3.55,"h":0.02,"id":"waiting","kind":"zone","label":"Waiting · 20 seats + 2 wheelchair spaces","source":"assumed","w":8.05,"x":0.2,"z":11.25},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-0","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":0.5,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-1","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":1.1800000000000002,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-2","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":1.86,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-3","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":2.54,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-4","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.22,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-5","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.9000000000000004,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-6","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":4.58,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-7","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.260000000000001,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-8","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.94,"z":11.3},{"color":6589308,"d":0.55,"h":0.85,"id":"wait-0-9","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":6.62,"z":11.3},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-0","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":2.65,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-1","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.2,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-2","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":3.75,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-3","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":4.3,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-4","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":4.85,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-5","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.4,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-6","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":5.95,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-7","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":6.5,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-8","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":7.05,"z":14.15},{"color":6589308,"d":0.55,"facing":180,"h":0.85,"id":"wait-1-9","kind":"chair","label":"Waiting seat","note":"Client to confirm waiting seat dimensions, specification and placement.","source":"assumed","w":0.55,"x":7.6,"z":14.15},{"color":13098711,"d":1.4,"h":0.02,"id":"wheel-0","kind":"zone","label":"Wheelchair","note":"Keep clear. Client to confirm wheelchair bay dimensions and approach.","source":"assumed","symbol":"wheelchair","w":0.9,"x":0.5,"z":13.35},{"color":13098711,"d":1.4,"h":0.02,"id":"wheel-1","kind":"zone","label":"Wheelchair","note":"Keep clear. Client to confirm wheelchair bay dimensions and approach.","source":"assumed","symbol":"wheelchair","w":0.9,"x":1.65,"z":13.35},{"color":14410462,"d":2.8,"h":3,"id":"lobby-w","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":8.25,"z":12},{"color":14410462,"d":2.8,"h":3,"id":"lobby-e","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":0.15,"x":12.6,"z":12},{"color":14410462,"d":0.15,"h":3,"id":"lobby-n","kind":"wall","label":"Partition","note":"150 mm concept partition; verify construction.","source":"assumed","w":4.5,"x":8.25,"z":11.85},{"color":15002091,"d":2.8,"h":0.02,"id":"lobby","kind":"zone","label":"Wind lobby · wheelchair / buggy parking","source":"assumed","w":4.2,"x":8.4,"z":12},{"d":0.15,"doorType":"automatic","h":2.2,"id":"lobby-inner","kind":"door","label":"Lobby 1.8 m automatic","note":"Width represents assumed clear opening; frame and pocket detailing to follow.","source":"assumed","w":1.8,"x":9.6,"z":11.85},{"color":13098711,"d":1.4,"h":0.02,"id":"lobby-wc","kind":"zone","label":"Wheelchair park","source":"assumed","symbol":"wheelchair","w":0.85,"x":8.55,"z":13.35},{"color":14278618,"d":1.3,"h":0.02,"id":"lobby-buggy","kind":"zone","label":"Buggy park","source":"assumed","w":0.8,"x":11.6,"z":13.35},{"color":15261127,"d":2.1,"h":0.02,"id":"reception","kind":"zone","label":"Check-in + payment / one staff each","source":"assumed","w":3.5,"x":12.9,"z":11.4},{"color":11836537,"d":0.7,"h":1.05,"id":"reception-counter","kind":"counter","label":"Check-in / one receptionist","note":"Client to confirm reception / payment dimensions, specification and placement.","source":"assumed","w":1.35,"x":14,"z":12.6},{"d":0.7,"h":0.76,"id":"reception-low","kind":"counter","label":"Accessible counter","note":"Lower section; detail knee recess and hearing loop.","source":"assumed","w":1,"x":15.35,"z":12.6},{"d":0.55,"facing":-90,"h":0.85,"id":"recept-chair-0","kind":"chair","label":"Payment receptionist","note":"Client to confirm receptionist dimensions, specification and placement.","source":"assumed","w":0.55,"x":13.35,"z":12.05},{"d":0.55,"h":0.85,"id":"recept-chair-1","kind":"chair","label":"Check-in receptionist","note":"Client to confirm receptionist dimensions, specification and placement.","source":"assumed","w":0.55,"x":14.95,"z":11.8},{"color":15851979,"d":2.5,"h":0.02,"id":"kids","kind":"zone","label":"Kids corner","source":"assumed","w":3.3,"x":17.2,"z":12.1},{"d":0.7,"h":0.5,"id":"kids-table","kind":"round_table","label":"Kids table","note":"Client to confirm kids table dimensions, specification and placement.","source":"assumed","w":0.7,"x":18.4,"z":13},{"d":0.35,"h":0.3,"id":"kids-stool-0","kind":"stool","label":"Kids seat","note":"Client to confirm kids seat dimensions, specification and placement.","source":"assumed","w":0.35,"x":17.9,"z":13.2},{"d":0.35,"h":0.3,"id":"kids-stool-1","kind":"stool","label":"Kids seat","note":"Client to confirm kids seat dimensions, specification and placement.","source":"assumed","w":0.35,"x":19.2,"z":13.2},{"d":1.5,"h":0.8,"id":"kids-shelf","kind":"bookshelf","label":"Books / toys","note":"Client to confirm books / toys dimensions, specification and placement.","source":"assumed","w":0.35,"x":20.45,"z":12.4},{"color":15196113,"d":3.6,"h":0.02,"id":"linen","kind":"zone","label":"Linen / storage · 1.7 × 3.6","source":"assumed","w":1.7,"x":12.8,"z":0.2},{"d":0.15,"h":3,"id":"linen-s","kind":"wall","label":"Storage partition","source":"assumed","w":1.7,"x":12.8,"z":3.8},{"d":0.15,"doorType":"sliding","h":2.2,"id":"linen-door","kind":"door","label":"Storage 1.0 m clear","source":"assumed","w":1,"x":13,"z":3.8},{"d":2.2,"h":1.8,"id":"linen-shelves","kind":"shelf","label":"Linen shelving","note":"Client to confirm linen shelving dimensions, specification and placement.","source":"assumed","w":0.5,"x":14,"z":0.2},{"color":14410462,"d":2.4,"h":3,"id":"toilet-chase","kind":"block","label":"Plumbing service void","source":"assumed","w":0.3,"x":7.05,"z":7.05},{"d":1.55,"h":2.75,"id":"lobby-glass-w","kind":"window","label":"Glazed lobby screen","source":"assumed","w":0.15,"x":8.25,"y":0.1,"z":13.2},{"d":1.5,"h":2.75,"id":"lobby-glass-e","kind":"window","label":"Glazed lobby screen","source":"assumed","w":0.15,"x":12.6,"y":0.1,"z":13.25},{"d":0.15,"h":2.75,"id":"lobby-glass-nw","kind":"window","label":"Lobby glazing","source":"assumed","w":1.2,"x":8.4,"y":0.1,"z":11.85},{"d":0.15,"h":2.75,"id":"lobby-glass-ne","kind":"window","label":"Lobby glazing","source":"assumed","w":1.2,"x":11.4,"y":0.1,"z":11.85},{"d":1.1,"doorType":"sliding","h":2.2,"id":"waiting-door","kind":"door","label":"Waiting 1.1 m clear sliding","source":"assumed","w":0.15,"x":8.25,"z":12.1},{"d":1.5,"h":0.02,"id":"xray-approach","kind":"zone","label":"X-ray approach · 1.5 m clear","source":"assumed","tags":["passage"],"w":5.35,"x":11.3,"z":10.25},{"d":0.5,"facing":90,"h":0.85,"id":"c1-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":0.3,"z":2.37},{"d":0.35,"h":1.1,"id":"c1-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":1.65,"z":0.2},{"d":0.5,"facing":90,"h":0.85,"id":"c2-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":3.45,"z":2.37},{"d":0.35,"h":1.1,"id":"c2-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":4.8,"z":0.2},{"d":0.5,"facing":90,"h":0.85,"id":"c3-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":6.6,"z":2.37},{"d":0.35,"h":1.1,"id":"c3-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":7.95,"z":0.2},{"d":0.5,"facing":90,"h":0.85,"id":"c4-companion","kind":"chair","label":"Companion chair","note":"Client to confirm companion seating and usable approach with selected chairs.","source":"assumed","w":0.5,"x":9.75,"z":2.37},{"d":0.35,"h":1.1,"id":"c4-cabinet","kind":"cabinet","label":"Clinical cabinet","note":"Client to confirm storage capacity, lock and cleanable finish.","source":"assumed","w":0.7,"x":11.099999999999998,"z":0.2},{"d":0.65,"h":0.85,"id":"treat-trolley","kind":"platform_trolley","label":"Clinical trolley","note":"Platform trolley proxy; client to confirm clinical trolley model and parking position.","source":"assumed","w":0.55,"x":0.2,"z":8.05},{"d":0.4,"h":0.65,"id":"treat-cabinet-2","kind":"cabinet","label":"Treatment cabinet","note":"Client to confirm wall cabinet storage and mounting height.","source":"assumed","w":0.9,"x":1.3,"y":1.5,"z":7.05},{"color":12245203,"d":1.8,"h":1.95,"id":"treat-curtain","kind":"screen","label":"Privacy curtain between couches","note":"Screen proxy for retractable curtain; client to confirm privacy coverage and track arrangement.","source":"assumed","w":0.06,"x":2.15,"z":9.15},{"color":15059361,"d":1.3,"h":0.01,"id":"kids-rug","kind":"rug","label":"Kids corner rug","note":"Client to confirm washable, slip-resistant rug and cleaning policy.","source":"assumed","w":2.2,"x":17.6,"y":0.025,"z":12.75},{"d":0.4,"h":1.1,"id":"wait-plant-1","kind":"plant","label":"Waiting plant","note":"Client to confirm planting and infection-control/maintenance policy.","source":"assumed","w":0.4,"x":0.2,"z":11.95},{"d":0.4,"h":1.1,"id":"wait-plant-2","kind":"plant","label":"Waiting plant","note":"Client to confirm planting and maintenance; keep approach clear.","source":"assumed","w":0.4,"x":7.75,"z":11.3},{"d":0.4,"h":1.1,"id":"wait-water","kind":"machine","label":"Water dispenser","note":"Machine proxy; client to confirm dispenser, supply, drip tray and servicing.","source":"assumed","w":0.45,"x":7.75,"z":13.65},{"d":0.6,"h":1,"id":"reception-files","kind":"cabinet","label":"Lockable files cabinet","note":"Client to confirm records capacity, locks and staff access.","source":"assumed","w":0.4,"x":13.95,"z":11.4},{"d":0.3,"facing":-90,"h":0.4,"id":"reception-workstation-1","kind":"pos_terminal","label":"Payment staff workstation","note":"Terminal proxy for computer; client to confirm IT, cabling and both staff handling check-in/payment.","source":"assumed","w":0.25,"x":12.8,"y":0.76,"z":12.1},{"d":0.3,"h":0.4,"id":"reception-workstation-2","kind":"pos_terminal","label":"Check-in staff workstation","note":"Terminal proxy for computer; client to confirm IT, cabling and both staff handling check-in/payment.","source":"assumed","w":0.4,"x":14.85,"y":1.05,"z":12.65},{"d":0.25,"h":0.2,"id":"reception-payment","kind":"pos_terminal","label":"Separate payment terminal","note":"Client to confirm terminal reach, sharing by both receptionists and accessible counter knee recess.","source":"assumed","w":0.25,"x":12.8,"y":0.76,"z":12.75},{"d":1.2,"h":0.76,"id":"payment-sill","kind":"counter","label":"Payment / lowered counter","note":"Kept inside reception footprint. Knee recess, hearing loop and glazing details are outside this model.","source":"assumed","w":0.35,"x":12.75,"z":12.05},{"d":1.2,"h":1.3,"id":"payment-window","kind":"window","label":"Separate payment window","note":"Service opening near the exit, with lowered counter; no circulation through the window.","source":"assumed","w":0.15,"x":12.6,"y":0.76,"z":12.05}],"radius":0.24,"scaleConfirmed":false,"serviceSeconds":15,"speed":0.6,"task":"clean","width":21},"x":0,"z":0}],"links":[],"name":"Linden Street Clinic / UC-MED-01 B — separate payment","storyHeight":4},"createdAt":"2026-10-09T12:03:10.973Z","id":"323ea01e-0106-4655-9308-7880eabcb807","name":"UC-MED-01 B / 4 rooms / 7 staff / 20 paired seeds","scenario":{"agents":[{"appearance":"person","height":1.7,"id":"patient","radius":0.3,"speed":1}],"arrivals":[{"at":{"floor":"floor-1","point":{"x":10.5,"z":14.1}},"id":"morning-arrivals","mode":"poisson","processId":"morning","profile":[{"durationSeconds":3600,"ratePerHour":24},{"durationSeconds":3600,"ratePerHour":16},{"durationSeconds":1800,"ratePerHour":10}],"startSeconds":0}],"horizonSeconds":21600,"id":"UC-MED-B-morning","maxEvents":100000,"name":"B / current four-room clinic / 20 paired seeds","parameters":{},"processes":[{"agent":"patient","id":"morning","steps":[{"to":{"floor":"floor-1","object":"reception-counter","side":"south"},"type":"goto"},{"count":1,"processId":"checkin-wait","type":"call"},{"durationSeconds":{"max":180,"min":60,"random":"uniform"},"type":"delay"},{"resourceId":"checkin-receptionist","type":"release","units":1},{"to":{"floor":"floor-1","point":{"x":4.225,"z":13.025}},"type":"goto"},{"count":1,"processId":"doctor-wait","type":"call"},{"count":1,"processId":"consult-room","type":"call"},{"resourceId":"doctors","type":"release","units":1},{"count":1,"processId":"after-consult","type":"call"},{"to":{"floor":"floor-1","point":{"x":12.05,"z":12.65}},"type":"goto"},{"count":1,"processId":"payment-wait","type":"call"},{"durationSeconds":{"max":120,"min":60,"random":"uniform"},"type":"delay"},{"resourceId":"payment-receptionist","type":"release","units":1},{"to":{"floor":"floor-1","point":{"x":10.5,"z":14.1}},"type":"goto"},{"type":"finish"}]},{"id":"checkin-wait","steps":[{"resourceId":"checkin-receptionist","type":"acquire","units":1},{"type":"finish"}]},{"id":"payment-wait","steps":[{"resourceId":"payment-receptionist","type":"acquire","units":1},{"type":"finish"}]},{"id":"doctor-wait","steps":[{"resourceId":"doctors","type":"acquire","units":1},{"type":"finish"}]},{"id":"return-doctor-wait","steps":[{"resourceId":"doctors","type":"acquire","units":1},{"type":"finish"}]},{"id":"consult-room","steps":[{"by":"free","options":[{"processId":"consult1","resourceId":"room1"},{"processId":"consult2","resourceId":"room2"},{"processId":"consult3","resourceId":"room3"},{"processId":"consult4","resourceId":"room4"}],"type":"branch"}]},{"id":"review-room","steps":[{"by":"free","options":[{"processId":"review1","resourceId":"room1"},{"processId":"review2","resourceId":"room2"},{"processId":"review3","resourceId":"room3"},{"processId":"review4","resourceId":"room4"}],"type":"branch"}]},{"id":"after-consult","steps":[{"options":[{"processId":"none","weight":70},{"processId":"treatment","weight":20},{"processId":"imaging","weight":10}],"type":"branch"}]},{"id":"none","steps":[{"type":"finish"}]},{"id":"treatment","steps":[{"to":{"floor":"floor-1","point":{"x":4.225,"z":13.025}},"type":"goto"},{"resourceId":"nurses","type":"acquire","units":1},{"to":{"floor":"floor-1","point":{"x":2.2,"z":8.5}},"type":"goto"},{"durationSeconds":{"max":900,"min":300,"random":"uniform"},"type":"delay"},{"resourceId":"nurses","type":"release","units":1},{"type":"finish"}]},{"id":"imaging","steps":[{"to":{"floor":"floor-1","point":{"x":4.225,"z":13.025}},"type":"goto"},{"resourceId":"radiographer","type":"acquire","units":1},{"to":{"floor":"floor-1","point":{"x":19.4,"z":9.9}},"type":"goto"},{"durationSeconds":{"max":600,"min":300,"random":"uniform"},"type":"delay"},{"resourceId":"radiographer","type":"release","units":1},{"to":{"floor":"floor-1","point":{"x":4.225,"z":13.025}},"type":"goto"},{"count":1,"processId":"return-doctor-wait","type":"call"},{"count":1,"processId":"review-room","type":"call"},{"resourceId":"doctors","type":"release","units":1},{"type":"finish"}]},{"id":"consult1","steps":[{"resourceId":"room1","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c1"},"type":"goto"},{"durationSeconds":{"max":720,"min":240,"mode":420,"random":"triangular"},"type":"delay"},{"resourceId":"room1","type":"release","units":1},{"type":"finish"}]},{"id":"review1","steps":[{"resourceId":"room1","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c1"},"type":"goto"},{"durationSeconds":{"max":240,"min":120,"random":"uniform"},"type":"delay"},{"resourceId":"room1","type":"release","units":1},{"type":"finish"}]},{"id":"consult2","steps":[{"resourceId":"room2","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c2"},"type":"goto"},{"durationSeconds":{"max":720,"min":240,"mode":420,"random":"triangular"},"type":"delay"},{"resourceId":"room2","type":"release","units":1},{"type":"finish"}]},{"id":"review2","steps":[{"resourceId":"room2","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c2"},"type":"goto"},{"durationSeconds":{"max":240,"min":120,"random":"uniform"},"type":"delay"},{"resourceId":"room2","type":"release","units":1},{"type":"finish"}]},{"id":"consult3","steps":[{"resourceId":"room3","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c3"},"type":"goto"},{"durationSeconds":{"max":720,"min":240,"mode":420,"random":"triangular"},"type":"delay"},{"resourceId":"room3","type":"release","units":1},{"type":"finish"}]},{"id":"review3","steps":[{"resourceId":"room3","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c3"},"type":"goto"},{"durationSeconds":{"max":240,"min":120,"random":"uniform"},"type":"delay"},{"resourceId":"room3","type":"release","units":1},{"type":"finish"}]},{"id":"consult4","steps":[{"resourceId":"room4","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c4"},"type":"goto"},{"durationSeconds":{"max":720,"min":240,"mode":420,"random":"triangular"},"type":"delay"},{"resourceId":"room4","type":"release","units":1},{"type":"finish"}]},{"id":"review4","steps":[{"resourceId":"room4","type":"acquire","units":1},{"to":{"floor":"floor-1","object":"c4"},"type":"goto"},{"durationSeconds":{"max":240,"min":120,"random":"uniform"},"type":"delay"},{"resourceId":"room4","type":"release","units":1},{"type":"finish"}]}],"resources":[{"capacity":1,"id":"checkin-receptionist","name":"One check-in staff","role":"shared"},{"capacity":1,"id":"payment-receptionist","name":"One payment staff","role":"shared"},{"capacity":2,"id":"doctors","role":"shared"},{"capacity":2,"id":"nurses","role":"shared"},{"capacity":1,"id":"radiographer","role":"shared"},{"capacity":1,"id":"room1","role":"equipment"},{"capacity":1,"id":"room2","role":"equipment"},{"capacity":1,"id":"room3","role":"equipment"},{"capacity":1,"id":"room4","role":"equipment"},{"at":{"floor":"floor-1","object":"waiting"},"capacity":1000,"id":"waiting-demand","mode":"occupy","role":"zone"}],"runs":20,"seed":20261009,"traceLimit":10000,"version":5,"warmupSeconds":0},"scenarioHash":"fnv1a-c2384595","spaceHash":"fnv1a-9911c034"},"createdAt":"2026-10-09T12:05:10.118Z","id":"e1c7afaf-93af-4d9c-baea-350c4f28811b","result":{"baseline":{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"firstRunMetrics":{"arrived":46,"averageCompletedCycleSeconds":2544.6657567455227,"averageWaitSeconds":1615.5869800320652,"blocked":0,"completed":46,"inProgress":0,"maxCompletedCycleSeconds":6036.905326406996,"maxWaitSeconds":4432.330081181552,"pendingArrivals":0,"planned":46,"totalDistanceMeters":4324.831311199389,"totalWaitSeconds":74317.001081475,"waiting":0},"firstRunStatus":"completed","inputHash":"fnv1a-aff24d3e","scenarioId":"UC-MED-A-morning","spaceHash":"fnv1a-11feab13","warnings":[]},"candidate":{"assumptions":["Collisions, avoidance, following, passing and re-routing between agents (robots or people): only capacities and FIFO queues are computed.","Density-dependent slowing, crowd pressure and the physical length of a queue (waiting agents overlap at one point): pedestrian flow is counts and capacities, not crowd physics.","Optimised lift group control (destination dispatch, load weighing beyond a full car, zoning, parking) and lift jerk: lifts with cars follow the fixed collective rules of the process catalog, at constant speed unless the scenario gives an acceleration and seconds per person boarding and alighting; an elevator without cars has only capacity and travel speed, so its waits are close to a lower bound.","Door opening times (except the fixed door and dwell time of a lift car at each stop, and its seconds per person boarding and alighting when given), acceleration (except a lift car's, when the scenario gives one), turning and vehicle orientation (circular footprints only), loaded shape, charging, batteries and failures.","Evacuation, safety codes and legal compliance, lines of sight and cameras.","Human decision-making beyond weighted branches and branches by the state of the run, learning or adaptive behaviour, and time-of-day changes other than arrival rate profiles, scheduled times and shifts set in advance (capacities of resources and zones, and the units of a pool in service, from given times).","Cleaning quality (coverage comes from simulate_space).","Agreement with real robots or real sites: determinism guarantees reproducibility, not field accuracy; calibrate separately.","Two-station transport (simulate_space) and process-scenario transport are separate implementations, kept consistent only by a matching test on one shared case; other layouts can differ.","Random values, poisson arrivals, branches and random picks draw from each arrival's and each job's own stream of the seed: the same scenario and seed give the same result, and a job keeps its draws when other jobs or streams change. Times are seconds and distances are metres.","goto follows the shortest route on a roadmap for each agent class, with the same circle-versus-obstacle clearance as the robot engine; connections are edges with their length and speed.","Capacities on resources, zones and connections are strict FIFO queues without preemption; waiting agents stand at one point, and nothing else slows movement.","Zones do not steer routes. An occupy zone counts agents that stop inside it; a transit zone counts every agent inside it, passing or not.","A connection end or a place on a blocking part moves to the nearest point the agent fits within 3 m; that offset is not added to the connection length.","Finish returns every resource still held by that job. Resources held by unfinished or blocked jobs remain in use.","Average wait includes every arrived job and unfinished waits up to the observation endpoint. Cycle averages include completed jobs only.","Repeated runs use seeds seed, seed + 1, …. Jobs, timelines, traces, series and warnings are the first run’s; replicates give every measure per run and spread its minimum, mean, maximum and population standard deviation."],"firstRunMetrics":{"arrived":46,"averageCompletedCycleSeconds":2537.220538883585,"averageWaitSeconds":1625.729627751855,"blocked":0,"completed":46,"inProgress":0,"maxCompletedCycleSeconds":5991.795999207726,"maxWaitSeconds":4398.848365443759,"pendingArrivals":0,"planned":46,"totalDistanceMeters":3515.789494439939,"totalWaitSeconds":74783.56287658533,"waiting":0},"firstRunStatus":"completed","inputHash":"fnv1a-b0c2596c","scenarioId":"UC-MED-B-morning","spaceHash":"fnv1a-9911c034","warnings":[]},"engineVersion":"process-des-2.7.0","horizonSeconds":21600,"measures":[{"baseline":43.35,"candidate":43.35,"delta":0,"id":"run.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":null,"scope":"run","unit":"count"},{"baseline":43.35,"candidate":43.35,"delta":0,"id":"run.arrived","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":null,"scope":"run","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"run.blocked","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":null,"scope":"run","unit":"count"},{"baseline":11868.578605942463,"candidate":11881.274171057164,"delta":12.695565114701731,"id":"run.completion.all","interval":{"high":51.506113945827515,"low":-26.114983716424057,"withinNoise":true},"paired":{"max":110.84655126680991,"mean":12.695565114701731,"min":-236.3599985205874,"n":20,"stdev":80.82716688878544},"ref":null,"scope":"run","unit":"s"},{"baseline":2854.2932563953214,"candidate":2857.2482302390945,"delta":2.954973843772689,"id":"run.cycle.mean","interval":{"high":13.364472367035209,"low":-7.454524679489832,"withinNoise":true},"paired":{"max":45.894664951896175,"mean":2.954973843772689,"min":-28.519320608077805,"n":20,"stdev":21.678906887643194},"ref":null,"scope":"run","unit":"s"},{"baseline":1902.16270762074,"candidate":1922.536314288935,"delta":20.373606668194753,"id":"run.wait.mean","interval":{"high":30.77435012764803,"low":9.972863208741476,"withinNoise":false},"paired":{"max":63.498071766449584,"mean":20.373606668194753,"min":-10.89197803235561,"n":20,"stdev":21.660673520042266},"ref":null,"scope":"run","unit":"s"},{"baseline":4101.452453229362,"candidate":3345.4639028562538,"delta":-755.9885503731073,"id":"run.distance.total","interval":{"high":-689.2960843409162,"low":-822.6810164052984,"withinNoise":false},"paired":{"max":-462.9814242453499,"mean":-755.9885503731073,"min":-1152.975019527239,"n":20,"stdev":138.89427602954615},"ref":null,"scope":"run","unit":"m"},{"baseline":7.225,"candidate":7.225,"delta":0,"id":"run.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":null,"scope":"run","unit":"per_hour"},{"baseline":6.000446126224175,"candidate":6.006607276731262,"delta":0.006161150507085322,"id":"run.wip.mean","interval":{"high":0.02682811338609503,"low":-0.014505812371924387,"withinNoise":true},"paired":{"max":0.0998633913305147,"mean":0.006161150507085322,"min":-0.054913051743723074,"n":20,"stdev":0.04304118617272325},"ref":null,"scope":"run","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"run.wip.end","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":null,"scope":"run","unit":"count"},{"baseline":0.21275376189386913,"candidate":null,"delta":null,"id":"resource.utilization","paired":null,"reason":"missing_value","ref":"receptionists","scope":"resource","unit":"ratio"},{"baseline":2.75,"candidate":null,"delta":null,"id":"resource.queue.max","paired":null,"reason":"missing_value","ref":"receptionists","scope":"resource","unit":"count"},{"baseline":14.094949757099878,"candidate":null,"delta":null,"id":"resource.wait.mean","paired":null,"reason":"missing_value","ref":"receptionists","scope":"resource","unit":"s"},{"baseline":137.97755449487013,"candidate":null,"delta":null,"id":"resource.wait.max","paired":null,"reason":"missing_value","ref":"receptionists","scope":"resource","unit":"s"},{"baseline":0.5056519563426073,"candidate":0.5056665668460949,"delta":0.00001461050348769144,"id":"resource.utilization","interval":{"high":0.00006704941889620418,"low":-0.00003782841192082131,"withinNoise":true},"paired":{"max":0.00029221006975321817,"mean":0.00001461050348769144,"min":-0.00022664238558423033,"n":20,"stdev":0.00010920971475135452},"ref":"doctors","scope":"resource","unit":"ratio"},{"baseline":14.1,"candidate":13.7,"delta":-0.4,"id":"resource.queue.max","interval":{"high":-0.04715066134471263,"low":-0.7528493386552875,"withinNoise":false},"paired":{"max":0,"mean":-0.4,"min":-3,"n":20,"stdev":0.7348469228349535},"ref":"doctors","scope":"resource","unit":"count"},{"baseline":1707.4678282977934,"candidate":1628.9317535410103,"delta":-78.53607475678226,"id":"resource.wait.mean","interval":{"high":-45.94034423110681,"low":-111.13180528245769,"withinNoise":false},"paired":{"max":0.6301574534675183,"mean":-78.53607475678226,"min":-287.2878752704414,"n":20,"stdev":67.88413538093796},"ref":"doctors","scope":"resource","unit":"s"},{"baseline":3151.592335026504,"candidate":3073.3391979540465,"delta":-78.25313707245752,"id":"resource.wait.max","interval":{"high":-27.788299338333523,"low":-128.7179748065815,"withinNoise":false},"paired":{"max":107.2130112786117,"mean":-78.25313707245752,"min":-396.8736838865734,"n":20,"stdev":105.09848441721195},"ref":"doctors","scope":"resource","unit":"s"},{"baseline":0.12189310230216592,"candidate":0.12189310230216592,"delta":1.734723475976807e-18,"id":"resource.utilization","interval":{"high":1.1032555691779651e-17,"low":-7.563108739826037e-18,"withinNoise":true},"paired":{"max":4.163336342344337e-17,"mean":1.734723475976807e-18,"min":-4.163336342344337e-17,"n":20,"stdev":1.936374153018703e-17},"ref":"nurses","scope":"resource","unit":"ratio"},{"baseline":0.3,"candidate":0.35,"delta":0.05,"id":"resource.queue.max","interval":{"high":0.23441168582392236,"low":-0.13441168582392238,"withinNoise":true},"paired":{"max":1,"mean":0.05,"min":-1,"n":20,"stdev":0.3840572873934304},"ref":"nurses","scope":"resource","unit":"count"},{"baseline":7.841542485322693,"candidate":9.261212791131914,"delta":1.419670305809222,"id":"resource.wait.mean","interval":{"high":5.020903242463552,"low":-2.1815626308451077,"withinNoise":true},"paired":{"max":23.721039132659573,"mean":1.419670305809222,"min":-17.43273289921688,"n":20,"stdev":7.499957211192754},"ref":"nurses","scope":"resource","unit":"s"},{"baseline":67.15597488889667,"candidate":75.44618219433148,"delta":8.29020730543482,"id":"resource.wait.max","interval":{"high":41.166118396497296,"low":-24.585703785627658,"withinNoise":true},"paired":{"max":213.48935219393616,"mean":8.29020730543482,"min":-191.7600618913857,"n":20,"stdev":68.46764172134228},"ref":"nurses","scope":"resource","unit":"s"},{"baseline":0.08711011776053709,"candidate":0.08711011776053709,"delta":-7.979727989493312e-18,"id":"resource.utilization","interval":{"high":9.373791228785891e-18,"low":-2.5333247207772515e-17,"withinNoise":true},"paired":{"max":5.551115123125783e-17,"mean":-7.979727989493312e-18,"min":-8.326672684688674e-17,"n":20,"stdev":3.614058126482083e-17},"ref":"radiographer","scope":"resource","unit":"ratio"},{"baseline":0.45,"candidate":0.4,"delta":-0.05,"id":"resource.queue.max","interval":{"high":0.05464999999999999,"low":-0.15465,"withinNoise":true},"paired":{"max":0,"mean":-0.05,"min":-1,"n":20,"stdev":0.21794494717703367},"ref":"radiographer","scope":"resource","unit":"count"},{"baseline":23.363559052288522,"candidate":18.692492503393797,"delta":null,"id":"resource.wait.mean","paired":null,"reason":"missing_value","ref":"radiographer","scope":"resource","unit":"s"},{"baseline":81.94981534115564,"candidate":72.7416662289563,"delta":-9.208149112199328,"id":"resource.wait.max","interval":{"high":39.384490707225396,"low":-57.800788931624055,"withinNoise":true},"paired":{"max":248.62361295513938,"mean":-9.208149112199328,"min":-309.61625987520983,"n":20,"stdev":101.199429705085},"ref":"radiographer","scope":"resource","unit":"s"},{"baseline":0.2649691770995189,"candidate":0.2654073935533545,"delta":0.0004382164538356464,"id":"resource.utilization","interval":{"high":0.008278213054009312,"low":-0.00740178014633802,"withinNoise":true},"paired":{"max":0.04592824958350883,"mean":0.0004382164538356464,"min":-0.01926864118516941,"n":20,"stdev":0.016327641136101034},"ref":"room1","scope":"resource","unit":"ratio"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.queue.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room1","scope":"resource","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.mean","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room1","scope":"resource","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room1","scope":"resource","unit":"s"},{"baseline":0.25653721914499594,"candidate":0.25822529104225106,"delta":0.0016880718972552195,"id":"resource.utilization","interval":{"high":0.008225661553797033,"low":-0.004849517759286593,"withinNoise":true},"paired":{"max":0.026621281728954477,"mean":0.0016880718972552195,"min":-0.0322586654608622,"n":20,"stdev":0.013615237767416409},"ref":"room2","scope":"resource","unit":"ratio"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.queue.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room2","scope":"resource","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.mean","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room2","scope":"resource","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room2","scope":"resource","unit":"s"},{"baseline":0.2446198412868495,"candidate":0.24455027088001763,"delta":-0.00006957040683180788,"id":"resource.utilization","interval":{"high":0.006907318744226295,"low":-0.00704645955788991,"withinNoise":true},"paired":{"max":0.03673972146013252,"mean":-0.00006957040683180788,"min":-0.027186043846068042,"n":20,"stdev":0.01453012649295153},"ref":"room3","scope":"resource","unit":"ratio"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.queue.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room3","scope":"resource","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.mean","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room3","scope":"resource","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room3","scope":"resource","unit":"s"},{"baseline":0.24517767515385,"candidate":0.24315017821656623,"delta":-0.002027496937283714,"id":"resource.utilization","interval":{"high":0.006328351838088917,"low":-0.010383345712656346,"withinNoise":true},"paired":{"max":0.036730833667366625,"mean":-0.002027496937283714,"min":-0.027779967567048425,"n":20,"stdev":0.01740195910146077},"ref":"room4","scope":"resource","unit":"ratio"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.queue.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room4","scope":"resource","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.mean","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room4","scope":"resource","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"room4","scope":"resource","unit":"s"},{"baseline":0.004058564987951154,"candidate":0.00386753812328612,"delta":-0.000191026864665035,"id":"resource.utilization","interval":{"high":-0.00009249360490049219,"low":-0.0002895601244295778,"withinNoise":false},"paired":{"max":0.000001108610334804496,"mean":-0.000191026864665035,"min":-0.0009310255217097572,"n":20,"stdev":0.00020520617386110107},"ref":"waiting-demand","scope":"resource","unit":"ratio"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.queue.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"waiting-demand","scope":"resource","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.mean","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"waiting-demand","scope":"resource","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"resource.wait.max","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"waiting-demand","scope":"resource","unit":"s"},{"baseline":2854.2932563953214,"candidate":2857.2482302390945,"delta":2.954973843773007,"id":"agentClass.completion.mean","interval":{"high":13.36447236703549,"low":-7.454524679489475,"withinNoise":true},"paired":{"max":45.894664951897084,"mean":2.954973843773007,"min":-28.519320608077805,"n":20,"stdev":21.67890688764311},"ref":"patient","scope":"agentClass","unit":"s"},{"baseline":1,"candidate":1,"delta":0,"id":"agentClass.feasible","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"patient","scope":"agentClass","unit":"ratio"},{"baseline":4101.452453229362,"candidate":3345.4639028562538,"delta":-755.9885503731073,"id":"agentClass.distance.total","interval":{"high":-689.2960843409162,"low":-822.6810164052984,"withinNoise":false},"paired":{"max":-462.9814242453499,"mean":-755.9885503731073,"min":-1152.975019527239,"n":20,"stdev":138.89427602954615},"ref":"patient","scope":"agentClass","unit":"m"},{"baseline":null,"candidate":null,"delta":null,"id":"agentClass.utilization","paired":null,"reason":"missing_value","ref":"patient","scope":"agentClass","unit":"ratio"},{"baseline":14.3,"candidate":13.75,"delta":-0.55,"id":"zone.occupancy.peak","interval":{"high":-0.19470876827297678,"low":-0.9052912317270233,"withinNoise":false},"paired":{"max":0,"mean":-0.55,"min":-3,"n":20,"stdev":0.7399324293474373},"ref":"waiting-demand","scope":"zone","unit":"count"},{"baseline":4.058564987951154,"candidate":3.8675381232861206,"delta":-0.19102686466503496,"id":"zone.occupancy.mean","interval":{"high":-0.0924936049004922,"low":-0.28956012442957774,"withinNoise":false},"paired":{"max":0.0011086103348045828,"mean":-0.19102686466503496,"min":-0.9310255217097563,"n":20,"stdev":0.20520617386110093},"ref":"waiting-demand","scope":"zone","unit":"count"},{"baseline":7.225,"candidate":7.225,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"morning","scope":"process","unit":"per_hour"},{"baseline":43.35,"candidate":43.35,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"morning","scope":"process","unit":"count"},{"baseline":2854.2932563953214,"candidate":2857.2482302390945,"delta":2.954973843773007,"id":"process.cycle.mean","interval":{"high":13.36447236703549,"low":-7.454524679489475,"withinNoise":true},"paired":{"max":45.894664951897084,"mean":2.954973843773007,"min":-28.519320608077805,"n":20,"stdev":21.67890688764311},"ref":"morning","scope":"process","unit":"s"},{"baseline":43.35,"candidate":43.35,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"morning","scope":"process","unit":"count"},{"baseline":2854.293256395322,"candidate":2857.248230239095,"delta":2.954973843772609,"id":"process.time.mean","interval":{"high":13.364472367035209,"low":-7.45452467948999,"withinNoise":true},"paired":{"max":45.89466495189754,"mean":2.954973843772609,"min":-28.519320608077805,"n":20,"stdev":21.678906887643357},"ref":"morning","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"checkin-wait","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"checkin-wait","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"checkin-wait","scope":"process","unit":"s"},{"baseline":43.35,"candidate":43.35,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"checkin-wait","scope":"process","unit":"count"},{"baseline":18.705477380671418,"candidate":122.93101026935054,"delta":104.22553288867907,"id":"process.time.mean","interval":{"high":137.1356328873578,"low":71.31543289000035,"withinNoise":false},"paired":{"max":313.4945486397693,"mean":104.22553288867907,"min":29.377863259528482,"n":20,"stdev":68.53884382038154},"ref":"checkin-wait","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"payment-wait","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"payment-wait","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"payment-wait","scope":"process","unit":"s"},{"baseline":43.35,"candidate":43.35,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"payment-wait","scope":"process","unit":"count"},{"baseline":9.484422133528332,"candidate":11.312732338517694,"delta":1.8283102049893614,"id":"process.time.mean","interval":{"high":4.366852259871348,"low":-0.7102318498926252,"withinNoise":true},"paired":{"max":11.575134608166447,"mean":1.8283102049893614,"min":-14.114244487233348,"n":20,"stdev":5.286788476425544},"ref":"payment-wait","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"doctor-wait","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"doctor-wait","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"doctor-wait","scope":"process","unit":"s"},{"baseline":43.35,"candidate":43.35,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"doctor-wait","scope":"process","unit":"count"},{"baseline":1698.3705860772639,"candidate":1619.499102056592,"delta":-78.8714840206718,"id":"process.time.mean","interval":{"high":-47.80229344552859,"low":-109.94067459581501,"withinNoise":false},"paired":{"max":0.647188735993609,"mean":-78.8714840206718,"min":-266.0794771293372,"n":20,"stdev":64.70495077623296},"ref":"doctor-wait","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"return-doctor-wait","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"return-doctor-wait","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"return-doctor-wait","scope":"process","unit":"s"},{"baseline":4.2,"candidate":4.2,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"return-doctor-wait","scope":"process","unit":"count"},{"baseline":1757.5501521727665,"candidate":1682.566467774842,"delta":null,"id":"process.time.mean","paired":null,"reason":"missing_value","ref":"return-doctor-wait","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult-room","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult-room","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"consult-room","scope":"process","unit":"s"},{"baseline":43.35,"candidate":43.35,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult-room","scope":"process","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"process.time.mean","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult-room","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review-room","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review-room","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"review-room","scope":"process","unit":"s"},{"baseline":4.2,"candidate":4.2,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review-room","scope":"process","unit":"count"},{"baseline":0,"candidate":0,"delta":null,"id":"process.time.mean","paired":null,"reason":"missing_value","ref":"review-room","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"after-consult","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"after-consult","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"after-consult","scope":"process","unit":"s"},{"baseline":43.35,"candidate":43.35,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"after-consult","scope":"process","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"process.time.mean","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"after-consult","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"none","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"none","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"none","scope":"process","unit":"s"},{"baseline":30.6,"candidate":30.6,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"none","scope":"process","unit":"count"},{"baseline":0,"candidate":0,"delta":0,"id":"process.time.mean","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"none","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"treatment","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"treatment","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"treatment","scope":"process","unit":"s"},{"baseline":8.55,"candidate":8.55,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"treatment","scope":"process","unit":"count"},{"baseline":640.310990876971,"candidate":641.7841718096789,"delta":1.4731809327078111,"id":"process.time.mean","interval":{"high":5.068282317244218,"low":-2.1219204518285957,"withinNoise":true},"paired":{"max":23.721039132659484,"mean":1.4731809327078111,"min":-17.748711457362106,"n":20,"stdev":7.487187590529077},"ref":"treatment","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"imaging","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"imaging","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"imaging","scope":"process","unit":"s"},{"baseline":4.2,"candidate":4.2,"delta":0,"id":"process.entered","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"imaging","scope":"process","unit":"count"},{"baseline":2456.5602653659553,"candidate":2377.184386883245,"delta":null,"id":"process.time.mean","paired":null,"reason":"missing_value","ref":"imaging","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult1","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult1","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"consult1","scope":"process","unit":"s"},{"baseline":11.1,"candidate":11.25,"delta":0.15,"id":"process.entered","interval":{"high":0.3790251247421511,"low":-0.07902512474215112,"withinNoise":true},"paired":{"max":1,"mean":0.15,"min":-1,"n":20,"stdev":0.47696960070847283},"ref":"consult1","scope":"process","unit":"count"},{"baseline":491.82388144918406,"candidate":486.16937591001425,"delta":-5.65450553916998,"id":"process.time.mean","interval":{"high":5.389431375396306,"low":-16.698442453736266,"withinNoise":true},"paired":{"max":42.6186763530024,"mean":-5.65450553916998,"min":-45.793420169411945,"n":20,"stdev":23.000193477989892},"ref":"consult1","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review1","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review1","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"review1","scope":"process","unit":"s"},{"baseline":1.3,"candidate":1.25,"delta":-0.05,"id":"process.entered","interval":{"high":0.2712099188594341,"low":-0.37120991885943405,"withinNoise":true},"paired":{"max":1,"mean":-0.05,"min":-1,"n":20,"stdev":0.6689544080129827},"ref":"review1","scope":"process","unit":"count"},{"baseline":201.82331141141165,"candidate":196.78048998225924,"delta":null,"id":"process.time.mean","paired":null,"reason":"missing_value","ref":"review1","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult2","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult2","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"consult2","scope":"process","unit":"s"},{"baseline":10.95,"candidate":10.8,"delta":-0.15,"id":"process.entered","interval":{"high":0.07902512474215112,"low":-0.3790251247421511,"withinNoise":true},"paired":{"max":1,"mean":-0.15,"min":-1,"n":20,"stdev":0.47696960070847283},"ref":"consult2","scope":"process","unit":"count"},{"baseline":486.8743044873353,"candidate":495.5417374234279,"delta":8.667432936092627,"id":"process.time.mean","interval":{"high":18.66727946343663,"low":-1.3324135912513757,"withinNoise":true},"paired":{"max":58.00680129041177,"mean":8.667432936092627,"min":-31.893111341422582,"n":20,"stdev":20.825762285527304},"ref":"consult2","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review2","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review2","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"review2","scope":"process","unit":"s"},{"baseline":1.05,"candidate":1.2,"delta":0.15,"id":"process.entered","interval":{"high":0.3790251247421511,"low":-0.07902512474215112,"withinNoise":true},"paired":{"max":1,"mean":0.15,"min":-1,"n":20,"stdev":0.4769696007084728},"ref":"review2","scope":"process","unit":"count"},{"baseline":184.7620111022373,"candidate":187.29471396873691,"delta":null,"id":"process.time.mean","paired":null,"reason":"missing_value","ref":"review2","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult3","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult3","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"consult3","scope":"process","unit":"s"},{"baseline":10.75,"candidate":10.8,"delta":0.05,"id":"process.entered","interval":{"high":0.28888013152382963,"low":-0.1888801315238296,"withinNoise":true},"paired":{"max":1,"mean":0.05,"min":-1,"n":20,"stdev":0.4974937185533101},"ref":"consult3","scope":"process","unit":"count"},{"baseline":478.67336161241644,"candidate":477.89341222730883,"delta":-0.779949385107642,"id":"process.time.mean","interval":{"high":9.83320946603073,"low":-11.393108236246015,"withinNoise":true},"paired":{"max":49.96571980304384,"mean":-0.779949385107642,"min":-53.48431561060221,"n":20,"stdev":22.103051554637553},"ref":"consult3","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review3","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review3","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"review3","scope":"process","unit":"s"},{"baseline":0.85,"candidate":0.75,"delta":-0.1,"id":"process.entered","interval":{"high":0.10929999999999992,"low":-0.3092999999999999,"withinNoise":true},"paired":{"max":1,"mean":-0.1,"min":-1,"n":20,"stdev":0.4358898943540672},"ref":"review3","scope":"process","unit":"count"},{"baseline":185.36031223030113,"candidate":184.620913207162,"delta":null,"id":"process.time.mean","paired":null,"reason":"missing_value","ref":"review3","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult4","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"consult4","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"consult4","scope":"process","unit":"s"},{"baseline":10.55,"candidate":10.5,"delta":-0.05,"id":"process.entered","interval":{"high":0.1888801315238296,"low":-0.28888013152382963,"withinNoise":true},"paired":{"max":1,"mean":-0.05,"min":-1,"n":20,"stdev":0.4974937185533101},"ref":"consult4","scope":"process","unit":"count"},{"baseline":483.83537011202844,"candidate":482.29680036886367,"delta":-1.5385697431647856,"id":"process.time.mean","interval":{"high":12.098299117589953,"low":-15.175438603919524,"withinNoise":true},"paired":{"max":52.19349577389403,"mean":-1.5385697431647856,"min":-60.83523847741992,"n":20,"stdev":28.4002547875521},"ref":"consult4","scope":"process","unit":"s"},{"baseline":0,"candidate":0,"delta":0,"id":"process.throughput.per_hour","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review4","scope":"process","unit":"per_hour"},{"baseline":0,"candidate":0,"delta":0,"id":"process.count.completed","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"review4","scope":"process","unit":"count"},{"baseline":null,"candidate":null,"delta":null,"id":"process.cycle.mean","paired":null,"reason":"missing_value","ref":"review4","scope":"process","unit":"s"},{"baseline":1,"candidate":1,"delta":0,"id":"process.entered","interval":{"high":0.21473726909142274,"low":-0.21473726909142274,"withinNoise":true},"paired":{"max":1,"mean":0,"min":-1,"n":20,"stdev":0.4472135954999579},"ref":"review4","scope":"process","unit":"count"},{"baseline":197.22031670264528,"candidate":202.35837921361343,"delta":null,"id":"process.time.mean","paired":null,"reason":"missing_value","ref":"review4","scope":"process","unit":"s"},{"baseline":11.1,"candidate":11.25,"delta":0.15,"id":"branch.chosen","interval":{"high":0.3790251247421511,"low":-0.07902512474215112,"withinNoise":true},"paired":{"max":1,"mean":0.15,"min":-1,"n":20,"stdev":0.47696960070847283},"ref":"consult-room/#1>consult1","scope":"branch","unit":"count"},{"baseline":10.95,"candidate":10.8,"delta":-0.15,"id":"branch.chosen","interval":{"high":0.07902512474215112,"low":-0.3790251247421511,"withinNoise":true},"paired":{"max":1,"mean":-0.15,"min":-1,"n":20,"stdev":0.47696960070847283},"ref":"consult-room/#1>consult2","scope":"branch","unit":"count"},{"baseline":10.75,"candidate":10.8,"delta":0.05,"id":"branch.chosen","interval":{"high":0.28888013152382963,"low":-0.1888801315238296,"withinNoise":true},"paired":{"max":1,"mean":0.05,"min":-1,"n":20,"stdev":0.4974937185533101},"ref":"consult-room/#1>consult3","scope":"branch","unit":"count"},{"baseline":10.55,"candidate":10.5,"delta":-0.05,"id":"branch.chosen","interval":{"high":0.1888801315238296,"low":-0.28888013152382963,"withinNoise":true},"paired":{"max":1,"mean":-0.05,"min":-1,"n":20,"stdev":0.4974937185533101},"ref":"consult-room/#1>consult4","scope":"branch","unit":"count"},{"baseline":1.3,"candidate":1.25,"delta":-0.05,"id":"branch.chosen","interval":{"high":0.2712099188594341,"low":-0.37120991885943405,"withinNoise":true},"paired":{"max":1,"mean":-0.05,"min":-1,"n":20,"stdev":0.6689544080129827},"ref":"review-room/#1>review1","scope":"branch","unit":"count"},{"baseline":1.05,"candidate":1.2,"delta":0.15,"id":"branch.chosen","interval":{"high":0.3790251247421511,"low":-0.07902512474215112,"withinNoise":true},"paired":{"max":1,"mean":0.15,"min":-1,"n":20,"stdev":0.4769696007084728},"ref":"review-room/#1>review2","scope":"branch","unit":"count"},{"baseline":0.85,"candidate":0.75,"delta":-0.1,"id":"branch.chosen","interval":{"high":0.10929999999999992,"low":-0.3092999999999999,"withinNoise":true},"paired":{"max":1,"mean":-0.1,"min":-1,"n":20,"stdev":0.4358898943540672},"ref":"review-room/#1>review3","scope":"branch","unit":"count"},{"baseline":1,"candidate":1,"delta":0,"id":"branch.chosen","interval":{"high":0.21473726909142274,"low":-0.21473726909142274,"withinNoise":true},"paired":{"max":1,"mean":0,"min":-1,"n":20,"stdev":0.4472135954999579},"ref":"review-room/#1>review4","scope":"branch","unit":"count"},{"baseline":30.6,"candidate":30.6,"delta":0,"id":"branch.chosen","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"after-consult/#1>none","scope":"branch","unit":"count"},{"baseline":8.55,"candidate":8.55,"delta":0,"id":"branch.chosen","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"after-consult/#1>treatment","scope":"branch","unit":"count"},{"baseline":4.2,"candidate":4.2,"delta":0,"id":"branch.chosen","interval":{"high":0,"low":0,"withinNoise":true},"paired":{"max":0,"mean":0,"min":0,"n":20,"stdev":0},"ref":"after-consult/#1>imaging","scope":"branch","unit":"count"},{"baseline":null,"candidate":0.24430847037628864,"delta":null,"id":"resource.utilization","paired":null,"reason":"missing_value","ref":"checkin-receptionist","scope":"resource","unit":"ratio"},{"baseline":null,"candidate":4,"delta":null,"id":"resource.queue.max","paired":null,"reason":"missing_value","ref":"checkin-receptionist","scope":"resource","unit":"count"},{"baseline":null,"candidate":122.93101026935054,"delta":null,"id":"resource.wait.mean","paired":null,"reason":"missing_value","ref":"checkin-receptionist","scope":"resource","unit":"s"},{"baseline":null,"candidate":443.4275247327797,"delta":null,"id":"resource.wait.max","paired":null,"reason":"missing_value","ref":"checkin-receptionist","scope":"resource","unit":"s"},{"baseline":null,"candidate":0.18119905341144965,"delta":null,"id":"resource.utilization","paired":null,"reason":"missing_value","ref":"payment-receptionist","scope":"resource","unit":"ratio"},{"baseline":null,"candidate":1.55,"delta":null,"id":"resource.queue.max","paired":null,"reason":"missing_value","ref":"payment-receptionist","scope":"resource","unit":"count"},{"baseline":null,"candidate":11.312732338517694,"delta":null,"id":"resource.wait.mean","paired":null,"reason":"missing_value","ref":"payment-receptionist","scope":"resource","unit":"s"},{"baseline":null,"candidate":110.41395991628951,"delta":null,"id":"resource.wait.max","paired":null,"reason":"missing_value","ref":"payment-receptionist","scope":"resource","unit":"s"}],"runs":20,"seed":20261009,"series":[{"baseline":{"sliceSeconds":180,"values":[1,3,5,5,9,8,8,8,8,6,8,10,11,11,11,10,10,11,11,12,11,11,13,14,15,15,15,15,15,14,14,13,13,13,13,14,14,13,15,15,13,15,15,14,13,13,12,13,13,13,13,12,12,10,10,9,8,8,7,5,4,4,2,2,2,2,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"candidate":{"sliceSeconds":180,"values":[1,3,5,5,9,8,8,8,8,6,8,9,10,10,10,9,10,12,12,12,12,11,13,14,15,15,15,16,16,14,14,13,12,12,13,14,14,13,15,15,13,15,15,14,13,13,12,13,13,13,13,13,11,11,10,9,9,8,8,6,4,4,2,2,2,2,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"key":"run.wip","kind":"run.wip","reads":"most","ref":null},{"baseline":{"sliceSeconds":180,"values":[0,0,1,3,3,4,4,4,4,2,2,3,5,6,6,6,7,8,9,8,9,8,8,9,9,11,12,12,12,12,11,10,10,9,9,9,10,9,10,11,10,10,10,9,9,10,9,10,10,9,9,9,8,7,6,6,5,4,3,2,2,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"candidate":{"sliceSeconds":180,"values":[0,0,1,2,2,2,3,4,4,2,2,3,3,4,4,5,6,6,8,8,8,8,8,8,9,10,11,12,12,12,12,10,10,9,9,9,9,9,10,10,10,10,10,9,9,10,9,10,10,9,9,9,8,7,6,6,5,4,3,2,2,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"key":"resource.waiting:doctors","kind":"resource.waiting","reads":"most","ref":"doctors"},{"baseline":{"sliceSeconds":180,"values":[0,0,0,1,3,1,4,4,4,2,2,0,3,4,6,6,7,5,8,8,7,8,7,8,9,9,11,12,12,12,11,10,10,8,9,7,8,9,9,10,10,10,10,9,9,9,9,8,9,9,9,9,8,7,6,6,5,4,3,2,2,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"candidate":{"sliceSeconds":180,"values":[0,0,0,1,2,2,2,3,4,2,2,1,3,3,4,4,4,6,6,8,7,8,8,8,8,9,10,11,10,12,12,10,10,8,9,7,9,9,9,10,10,10,10,9,9,9,9,8,10,9,9,9,8,7,6,6,5,4,3,2,2,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"key":"zone.occupancy:waiting-demand","kind":"zone.occupancy","reads":"most","ref":"waiting-demand"},{"baseline":{"sliceSeconds":180,"values":[0,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"candidate":{"sliceSeconds":180,"values":[0,1,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"key":"resource.inUse:doctors","kind":"resource.inUse","reads":"most","ref":"doctors"},{"baseline":{"sliceSeconds":180,"values":[0,0,0,0,1,1,2,2,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,2,1,1,1,0,0,1,1,1,1,1,1,1,1,1,2,2,1,1,0,0,0,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"candidate":{"sliceSeconds":180,"values":[0,0,0,0,0,1,2,2,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,2,1,1,1,1,0,0,1,1,1,1,1,2,1,1,2,2,1,1,0,0,0,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"key":"resource.inUse:nurses","kind":"resource.inUse","reads":"most","ref":"nurses"},{"baseline":{"sliceSeconds":180,"values":[0,0,0,0,0,0,0,0,1,1,1,1,0,1,1,1,0,0,0,0,0,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"candidate":{"sliceSeconds":180,"values":[0,0,0,0,0,0,0,0,1,1,1,1,0,0,1,1,1,0,0,0,0,0,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]},"key":"resource.inUse:radiographer","kind":"resource.inUse","reads":"most","ref":"radiographer"}],"spaceHash":null,"warmupSeconds":0},"version":"process-comparison-1"}],"contentHash":"fnv1a-502ba6e6","id":"src-facbccd0-f1d3-4b37-9e10-c511f73c228f","modelRevision":"1a4b23fa-7bb2-4b6c-bcde-772d97e13473:29","projectName":"Linden Street Clinic / UC-MED-01 A — shared reception","spaceHash":"fnv1a-11feab13"},"warnings":["Drawing dimensioned-plan-plan is an inset of a 297 × 210 mm sheet at 1:150; use written dimensions or export the full-size drawing."]}