A hybrid multi-objective iterated local search heuristic for vehicle routing problem with time windows

The Vehicle Routing Problem with Time Windows (VRPTW) is a well known NP-Hard combinatorial optimization problem and it has received a lot of attention in the literature. In this problem, a fleet of identical vehicles must leave the depot, supply all costumers demands, and return to the depot, at minimum cost, without violating the capacity of the vehicles as well as the time window specified by each customer. This paper considers a VRPTW which aims to minimize two objective functions simultaneously: the total traveling distance and the imbalance in the distances traveled by the vehicles used. To obtain near Pareto-optimal solutions, we propose a hybrid heuristic based on Iterated Local Search, Variable Neighbourhood Descent with random neighbourhood ordering for solution improvement and Recombination of non-dominated solutions as used in genetic algorithms. The results obtained when solving a subset of Solomon´s benchmark problems show the good performance of the hybrid heuristic.