Start from the business hotel ground floor model (free sign-up; running a simulation is Pro). When check-in opens at 15:00, many guests line up at the front desk at once. In this article, in the members’ new model of a business hotel’s ground floor, your AI simulates an afternoon of check-ins. It ran the same tools an AI calls over MCP. Every figure and number below comes from that run.
1. The ground floor and the afternoon
- A ground floor of 16 × 12 m with the front desk of three stations and its line, two self check-in kiosks, the lobby and the lift hall.
check_spacefinds nothing.
The AI wrote this flow as a scenario (put_process_scenario).
- From 15:00 to 21:00, 110 parties come in at the entrance at random: 25 % from 15:00, then 20, 18, 15, 12 and 10 % an hour.
- A check-in takes 2 to 6 minutes at the front desk (most often 3), 1 to 3 minutes at a kiosk (most often 1.5). The front desk has as many staff as stations open.
- A party waits for a free station or kiosk, and when done goes to the lift hall.
- Every number and time is an assumption. Put in your own hotel’s check-in records.
2. The staff and the check-ins
The AI changed the front desk’s staff (1, 2, 3) and the check-ins (80, 110, 140 parties) (sweep_process_parameters), 20 runs each. Everyone checks in at the front desk. The longest wait is the mean over the 20 runs of each run’s longest.
| Check-ins | Front desk staff | Wait (mean) | Longest wait | In line (most) |
|---|---|---|---|---|
| 80 | 1 | 19.0 min | 40.9 min | 11.4 |
| 80 | 2 | 0.6 min | 5.0 min | 3.2 |
| 80 | 3 | 0.1 min | 2.2 min | 1.8 |
| 110 | 1 | 69.3 min | 112.9 min | 31.0 |
| 110 | 2 | 2.0 min | 10.3 min | 5.9 |
| 110 | 3 | 0.2 min | 3.2 min | 3.0 |
| 140 | 1 | 124.9 min | 205.1 min | 55.9 |
| 140 | 2 | 6.6 min | 20.3 min | 11.2 |
| 140 | 3 | 0.6 min | 5.3 min | 4.6 |
- With 110 parties, a party waits 69.3 minutes on average with one at the front desk, 2.0 with two and 0.2 with three.
- With 140 and two, the wait is 6.6 minutes on average and the longest 20.3.
3. Those who use the kiosks
With 110 parties, the AI changed the share who use the self check-in kiosks (none, 30 %, 60 %) and the front desk’s staff.
| Kiosks | Front desk staff | Front desk: wait (mean) | Front desk: longest wait | Kiosks: longest wait |
|---|---|---|---|---|
| None | 1 | 69.3 min | 112.9 min | — |
| None | 2 | 2.0 min | 10.3 min | — |
| 30 % | 1 | 17.2 min | 40.4 min | 0.1 min |
| 30 % | 2 | 0.7 min | 5.5 min | 0.1 min |
| 60 % | 1 | 2.6 min | 12.1 min | 1.0 min |
| 60 % | 2 | 0.2 min | 2.8 min | 1.0 min |
- With 60 % at the kiosks, even one at the front desk keeps the wait to 2.6 minutes on average: close to two staff with no kiosk use (2.0 minutes).
- The two kiosks, with 60 % using them, keep their longest wait to 1.0 minute.
4. How to use this
- Draw your hotel’s ground floor (from a PDF of the drawings or photos too), and tell the AI the check-ins each hour and how long the front desk and the kiosks take. The AI puts any kiosks on the drawing and checks them.
- Not checked: group arrivals, guests who take longer (other languages, payment), check-outs and phone calls at the desk, luggage, and the waits for the lifts.