Paper Title
APPLYING DEEP LEARNING TO ORDER BATCHING STRATEGY IN A PICK-AND-PASS WAREHOUSING SYSTEM

Abstract
Abstract - In recent years, due to the vigorous development of e-commerce online shopping, the type of orders tends to be small volume and diverse. Pick-and-pass warehousing system suitable for this type has play an essential role in supply chain. The order batching strategy can reduce the total distance of pickers via combining multiple orders into a batch, then picking items in one tour. Therefore, order batching strategy is one of the most common methods to improve the efficiency of order picking. However, most of the previous methods have primarily focused on the traditional picker-to-part warehouse system, and are difficult to implement in practice because of complexity. With the rapid development and evolution of artificial intelligence, the technology of deep learning is mature enough to be effectively applied in the warehouse environment with high complexity of orders. Therefore, this study is the first one to apply deep learning to develop order batching method in pick-and-pass warehousing system for quickly grouping orders with similar picking tour into batches. The results show that the order batching strategy proposed in this study can effectively shorten the total distance of picking and improve the efficiency in the pick-and-pass warehousing system. Keywords - Deep Learning, Picking system, Pick-and-Pass warehouse System, Order Batching