The order picking efficiency in robotic mobile fulfillment systems is not only determined by the order and rack processing sequences, but also by the product distribution on the racks. In this paper, we focus on long-term planning based on historical order data to identify the optimal distribution of SKUs on racks for order picking operation improvement. This product distribution problem is formulated as an integer linear program, with the goal of minimizing rack movements required to fulfill orders. We develop a solution procedure that takes into account both the affinity and frequency of products in the orders, as well as the corresponding performance of the order and rack sequencing. We evaluate the performance of the proposed method through comparisons with the theoretical optimal solution on manually generated small instances and with existing heuristic solutions on a large real-world order dataset provided by a major e-commerce company, demonstrating a significant improvement from using our algorithm.

中文翻译：

---

通过机器人移动履行系统中的存储分配优化来提高订单拣选效率

机器人移动履行系统中的订单拣选效率不仅取决于订单和货架处理顺序，还取决于货架上的产品分布。在本文中，我们重点关注基于历史订单数据的长期规划，以确定货架上 SKU 的最佳分布，以改进订单拣选操作。该产品分配问题被表述为整数线性程序，其目标是最大限度地减少履行订单所需的货架移动。我们开发了一个解决方案程序，该程序考虑了订单中产品的亲和力和频率，以及订单和机架排序的相应性能。我们通过与手动生成的小实例上的理论最优解决方案以及大型电子商务公司提供的大型现实世界订单数据集上的现有启发式解决方案进行比较来评估所提出方法的性能，证明使用我们的算法有显着的改进。