A hybrid metaheuristic for the Vehicle Routing Problem with Time Windows

Abstract

In this paper we propose to solve the Vehicle Routing Problem with Time Windows (VRPTW) using a hybrid metaheuristic. The VRPTW is a bi-objective optimization problem where both the number of vehicles and the distance of the travel to use should be minimized. Because it is often difficult to optimize both objectives, we propose an approach that optimizes the distance traveled by a fleet of vehicles. Such a strategy has been already used by several authors in the domain. Herein, an instance of VRPTW is considered as the composition of the Assignment Problem and a series of Traveling Salesman Problems with Time Windows (TSPTW). Both AP and TSPTW are solved by using an ant colony optimization system. Furthermore, in order to enhance the quality of the current solution, a large neighborhood search is introduced. Finally, a preliminary experimental part is presented where the proposed method is evaluated on a set of benchmark instances and its results are compared to the best results obtained by the methods available in the literature. Our preliminary results show that the proposed hybrid method remains competitive and it is able to reach new minimum distances for some tested instances.