A Multi-Objective Imperialist Competitive Algorithm for Vehicle Routing Problem in Cross-docking Networks with Time Windows

This study addresses the pickup and delivery problem for cross-docking strategy, in which shipments are allowed to be transferred from suppliers to retailers directly as well as through cross-docks. Usual models that investigate vehicle routing in cross-docking networks force all vehicles to stop at the cross-dock even if a shipment is about to a full truckload or the vehicle collects and delivers the same set of products. In order to eliminate unnecessary stops at the dock, and thus reduce transportation costs, the designed model tries to decide about the best approach to deliver orders to retailers in a tailored network. In such a system, two objectives are taken into account: minimization of the total transportation cost and minimization of the total earliness and tardiness of visiting retailers. In order to deal with this problem, three multi-objective algorithms are developed. An evolutionary algorithm based on multi objective imperialist competitive algorithm (MOICA) is proposed, and the associated results are compared with the results obtained by non-dominated sorting genetic algorithm (NSGA-II) and Pareto archived evolution strategy (PAES) in terms of some metrics. The computational results show the superiority of the proposed algorithm compared to other algorithms in some metrics.