Paper Title
APPLYING DEEP LEARNING TO THE ORDER PICKING BATCH PROBLEM OF A MULTI-RACK WAREHOUSING SYSTEM

Abstract
Abstract - With the increasing demands in logistics, the importance of warehouse management has been growing due to the large volume of orders. The efficiency of order picking systems is crucial in this context. The development of artificial intel-ligence has shown remarkable performance in improving efficiency and has been widely applied in various fields. Previous research on order picking has mostly focused on single-layer rack systems, neglecting considerations of rack height and vertical picking distances. Furthermore, the integration of order picking systems with deep learning techniques has been lacking. However, multi-layer rack systems can effectively utilize warehouse space and have become one of the mainstream warehousing configurations today. Therefore, this study aims to apply deep learning techniques to multi-layer rack warehouse systems, specifically focusing on order batching policy allocation based on picking routes. The research results demonstrate that using deep learning for order batch allocation can effectively reduce picking route distances and improve the order picking efficiency of multi-layer rack warehouse systems. Keywords - Picking Systems, Deep Learning, Multi-Layer Rack Systems, Order Batching Policy.