Project Details
Home/Project Details
Campus Shuttle
Project description
A shuttle system that knows where the passengers are waiting
The Challenge
Campus shuttle services run on a fixed route and a driver's judgement, while demand concentrates and moves: one stop is empty when the timetable says it should be busy, and another has forty people watching a full bus go past. The schedule and the need are two different shapes.
The cheapest improvement is not more buses — it is information. Any intervention has to assume the fleet is already fully used, and it has to work for drivers and passengers who are not going to install something complicated.
The Solution
Model demand from where passengers actually are rather than from the timetable, then test whether a demand-responsive route beats a fixed one on the same fleet. The work is a study first and a product second: the notebook establishes what the data says before anything is built around it.
Skills and deliverables
Features delivered
The hard part
Getting data at all. There is no telemetry on a campus shuttle, so the model is built from observed patterns and stated assumptions, and the study is explicit about which is which rather than presenting an estimate as a measurement.
The Impact
A research prototype: the demand model, the route comparison and the passenger-driver coordination concept, in notebooks that state their assumptions. It is a study and the record labels it as one.
