A memetic algorithm and a tabu search for the multi-compartment vehicle routing problem

A generalization of the traditional vehicle routing problem (VRP) is studied in this paper. Each customer may order several products, the vehicles are identical and have several compartments, each compartment being dedicated to one product. The demand of each customer for a product must be entirely delivered by one single vehicle. However, the different products required by a customer may be brought by several vehicles. Two algorithms to solve this problem are proposed: a memetic algorithm with a post-optimization phase based on path relinking, and a tabu search method. These algorithms are evaluated by adding compartments to classical VRP instances.