Elevator Dispatch Simulator
A cutaway office tower, a bank of elevators, and a growing crowd in the lobby. Passengers arrive in realistic traffic patterns — morning up-peak, lunchtime chaos, evening down-peak — and the dispatch algorithm decides which car answers which call. Race the classic collective control (the up-and-down sweep every old building uses) against nearest-car assignment and modern destination dispatch, where you key in your floor before boarding and the system groups passengers by destination. Live histograms track waiting time and time-to-destination, and queue lengths make bad algorithms visible at a glance. Tune the building — floors, cars, arrival rate — and find where each strategy breaks down.
Runs 100% in your browser — simulations are computed locally on your device.
Read the full guide to this tool
Notes
- Elevator dispatch is a queueing-theory problem: passengers arrive stochastically, service capacity is limited, and above a critical arrival rate queues grow without bound no matter the algorithm.
- Collective control (SCAN) is the elevator version of the disk-scheduling algorithm: sweep up answering every call, then sweep down — simple, fair, and wasteful in up-peak traffic.
- Destination dispatch gains its edge by batching: grouping passengers with the same destination cuts the number of stops per trip, which is why it dominates in tall buildings during rush hour.
- Runs 100% in your browser — simulations are computed locally on your device.