A competitive neural network algorithm for solving vehicle routing problem

This paper describes a self organization Neural Network algorithm for a class of Vehicle Routing Problems. Motivated by the outstanding performance of adaptive Neural Network approach in the Traveling Salesman Problem, we devised an algorithm to extend the domain of applicability of this approach to more complex problems. First, relevant adaptation is proposed to refine the model for the Multiple Traveling Salesman Problem. Then, an additional mechanism to satisfy further constraints are embodied into the algorithm. The effectiveness of the proposed algorithm is evaluated by considering a series of standard problems from the literature. The results show that the algorithm can yield solutions within a few percent of optimality.