Research On Order Batching Problem In Distribution Center

Order picking is one of the main processes in logistics centers, and there is about 60 percent of cost that is caused by order picking in logistics center. So the picking efficiency is necessary to be improved to reduce the cost in logistics center. Before picking a customer order, grouping some orders into one batch and then completing them during the same tour can significantly save the picking time and thus improve the efficiency of picking the goods. Even quite few picking time is saved, it leads to reduce cost in distribution center. The purpose of this paper is to study an effective method to group orders into batches and to minimize the total distance when completing all goods picking.Firstly, in a view of route choice, namely, first choosing one best route for each order from route pool, then grouping all orders which belong to the same route into batches, the minimum walking distance of picking all orders as the objective function is developed on the basis of summarizing the order batching algorithm.Secondly, a hybrid genetic algorithm and iterative local search algorithm is proposed to solve the model. Two steps are divided to address the model to get a best solution. Step 1 is to develop route pool for all orders. Step 2 is to group orders into batches by combining GA and ILS. GA is used to choose a best route for each order from route pool. ILS is adopted to minimize the number of batches for those orders that belong to the same route.Finally,8 different experiments are carried out to prove the performance of proposed GA-ILS algorithm with the consideration on the equipment capacity and the number of orders. From the perspective of computational efficiency and the degree of objective function optimization respectively, this paper compares three algorithms, namely, the proposed hybrid genetic algorithm and iterative local search algorithm (GA-ILS), first come first serve (FCFS) and the base genetic algorithm (base-GA). The results show that the proposed GA-ILS performs better in reducing walk distance than FCFS and base-GA, as well as improves the calculation efficiency comparing with base-GA. Therefore, the proposed GA-ILS can be potentially applied to practice.