On-demand night bus: pooling rides home around Windhagen
Which bus picks up which booking, in which order — so everyone gets home at night with short waits and few empty kilometres?
Drag to rotate, scroll or pinch to zoom. The schedule below plays the plan; hover a bar for details. Plan computed with a Timefold model.
The planning question
On a Saturday night, passengers book a ride home: from the central bus stop when the last trains arrive, from the village pub, to the last regular bus, or between villages. Three 8-seat minibuses serve 16 villages around Windhagen; travel times come from the real road network.
How we plan it
A Timefold pickup-and-delivery model: each booking is a pickup and a drop-off on the same bus, with seats, time windows and ride limits checked along the route.
What the plan shows
| KPI | Optimized plan | Comparison: booking order, one ride at a time |
|---|---|---|
| bookings served | 53 / 53 +13 | 40 / 53 |
| average wait | 2.2 min −2.7 | 4.9 min |
| bus kilometres | 159.4 km −68.8 | 228.2 km |
| shared rides | 68 % +68 | 0 % |
Decisions
- bus per booking
- order of pickups and drop-offs on each bus
- pickup and drop-off times
Constraints
- pickup within 15–20 minutes of the booked time
- maximum ride time (1.8 × the direct ride + 5 minutes)
- 8 seats per bus
- rides to the last bus arrive before it leaves
- buses back at the stop by 03:30
Objective
serve every booking with short waits and detours, few bus kilometres, many shared rides
Questions
Is this like a taxi?
No — the point is pooling: passengers going in the same direction share the bus, which serves more people with fewer kilometres.
Can new bookings come in during the night?
Yes. Rides on board stay fixed, the open part of the night is re-planned when a booking arrives.
More use cases
Your planning problem looks different?
Every operation has its own rules. Tell us about yours — we assess feasibility and expected impact in a free initial call.
Book a free call Write to us