Improving a Vehicle Routing Heuristic Through Genetic Search

The Vehicle Routing Problem with Time Windows (VRPTW) is currently the focus of very intensive research, and is used to model many realistic applications. The overall objective is to serve a set of customers at minimum cost with a fleet of vehicles of finite capacity housed at a central depot. The route of each vehicle starts and ends at the depot. Also, each customer has a known demand (i.e. some quantity of goods to be picked-up), as well as a time window or time interval that constrains the vehicle's arrival time. The time windows are defined within a scheduling horizon, so that each route starts and ends within the bounds of this horizon. When the horizon or the capacity is enlarged, a larger number of customers can be served by the same vehicle (and conversely).