Vehicle Routing and Scheduling

In this paper we propose a heuristic algorithm to solve the Vehicle Routing Problem with Time Windows. Its main feature is simplicity, since it combines the classical k-opt exchanges with an ad-hoc procedure to reduce the number of vehicles, and it alternates between two objective functions in order to escape local optima. The distinguishing features with respect to the other proposed algorithms is full determinism: no random choice is done. The algorithm has been tested on benchmark problems proving to be more effective than comparable algorithms.

[email protected],