Abstract: In the "part-to-picker" picking system which can pick multiple customer orders in parallel, each customer has multiple orders, and requires to receive commodities by the order sequence. To improve picking efficiency of this system, order scheduling and customer batching are optimized respectively, aiming at minimizing the number of "In-Out Stock" boxes. 0-1 integer programming model is established for scheduling orders of customers. Two different 0-1 integer programming models are established to solve customer batch optimization problems with multiple customers picking at the same time and the order scheduling optimization problem respectively. In addition, the seed algorithm and genetic algorithm are proposed for solving the customer batch problem. Numerical experiments are designed to evaluate the performance of proposed order scheduling model and algorithms of customer batching problem in different customer number, order number and stock keeping unit number. The experimental results show that the 0-1 integer programming model for optimizing order sequence can improve the efficiency by about 15%, which is effective. For customer batch optimization, 0-1 integer programming model, genetic algorithm and seed algorithm can improve the efficiency of the system with varying degrees, which can be applied for scenarios with different problem sizes and time requirements.

Key words: “part-to-picker” picking, customer batching, order scheduling, 0-1 integer programming, seed algorithm, genetic algorithm