Application of Simulated Annealing Based Method for Order Batching and Routing Problems in Logistics Warehouses

Abstract

Order picking is a research field that examines methods for retrieving items stored in a warehouse based on customer orders. It is the most labor-intensive process in a logistics warehouse, necessitating optimization for efficiency. Order batching and picker routing are key approaches for improving the efficiency of order picking operations in logistics warehouses. The former focuses on optimizing the grouping of multiple orders to enhance picking efficiency, while the latter aims to optimize the sequence in which all SKUs (Stock Keeping Units) that are requested in an order are picked. Both have been formulated as mathematical optimization problems for minimizing the total travel distance, and extensive research has been conducted on them. However, as the problem scale increases, obtaining an exact solution within a practical timeframe becomes infeasible, and even an approximate solution may be unattainable. Therefore, this study simultaneously considers the order batching problem and the picker routing problem in a logistics warehouse. A new optimization algorithm for batching and routing problem is proposed based on the simulated annealing method, which is a metaheuristic. The effectiveness of the proposed method was validated through extensive numerical experiments. In addition, batches obtained using the proposed method and a conventional method using four evaluation criteria to analyze which batches are superior. Furthermore, by first conducting the search based on aisle distance to enhance efficiency, then refining batches using shelf distance, better solutions were achieved for large-scale problems with shorter search times compared to models that rely solely on shelf distance.