عنوان مقاله [English]
Cross docking is a warehouse management concept in which items delivered to a warehouse by inbound trucks are immediately sorted out and reorganized based on customer demands and are routed and loaded into outbound trucks for delivery to customers without being held in inventory in the warehouse. If any item is to be held in storage, it is only for a brief period of time that is typically less than 24 hours. Based on this concept, inventory management cost, turn-around times for customer orders, and warehouse space requirements are reduced. In another definition, a cross dock is a consolidation point in a distribution network, where multiple smaller shipments can be merged with full truck loads to decrease the transportation costs. In cross-docking systems, the truck scheduling problem, which decides on the succession of inbound and outbound truck processing at the dock doors, is significantly important to guarantee a rapid turnover and on-time deliveries. The cost reduction of the cross-docking systems is proved by the successful implementation of several industries: the retail chain (Wal-Mart), the mailing companies (UPS), the automobile manufacturers (Toyota), and less-than-truckload providers. Cross-docking systems can be distinguished based on when the customer is assigned to the individual products. In pre-distribution cross-docking (Pre-C), the customer is assigned before the shipment leaves the supplier who takes care of preparation and sorting. On the other hand, in post-distribution cross-docking (Post-C), the allocation of goods to customers is done at the cross-dock. In this paper, the truck scheduling problem in pre-distribution cross-docking systems is studied and a multi-objective model based on Mixed Integer Programming is proposed. For solving the proposed model, three multi-objective genetic- based algorithms are developed: Non Dominated Sorting Genetic Algorithm-II (NSGA-II), Pareto Envelope based Selection Algorithm-II (PESA-II), and Strength Pareto Evolutionary Algorithm (SPEA-II). In order to evaluate the performance of the meta-heuristics, several numerical examples are randomly generated along with those presented in the literature. At last, the Pareto fronts of three algorithms are compared by three evaluation metrics
which contain: Mean Ideal Distance (MID), Spacing Metric (SM), and Quality Metric (QM). The results show that among these developed algorithms, the SPEA-II obtains the best performance based on all evaluation metrics.