Project Proposal by Martin Stacey
Vehicle Rental Simulation
Software |
Object oriented language |
Covers |
Discrete event simulation, possibly statistics and business modelling |
Skills Required |
Programming, ideally some interest in either cycling or the future of cars |
Challenge |
Conceptual Technical Programming |
Brief Description
Owning a car is an expensive proposition before you even start driving it, and modern cities are clogged up with cars parked on streets and in car parks, not doing anything. Some visionaries are looking at the rapid development of working driverless cars, and arguing that that they shouldn't just challenge human-driven taxis but also car ownership. If a car can get itself to where you need it, you don't need exclusive use of something you only make use of for an hour a day.
Some cities have experimented with civic bicycle rental as a resource for both locals and visitors - you can pick up and drop off bicycles when you need them in city centres. Payment mode and losses are issues, but some schemes have proved successful.
Your challenge is to build a simulation of a transport for rent system to see how well it would work given both random and predictable variations in usage, given different assumptions about total capacity (primarily number of vehicles), usage patterns, loss rates, payments, costs and the like.
The simulation should be designed so that a simple simulation including only a small number of factors influencing the behaviour of the elements is relatively straightforward to build, and additional realism and complexity can be added incrementally. The work should include systematic testing of how the simulation behaves under different sets of assumptions.
Back to