Design And Implementation Of Order Combination Picking System Based On Hierarchical Clustering

With the integration of Internet technology and e-commerce warehousing development,modern warehousing operation mode is gradually transformed from the traditional man-to-goods mode to the goods-to-man mode.The integration of new technologies has brought more new possibilities to warehouse operations and more challenges to logistics speed.This paper takes an e-commerce warehouse as the object,and through the investigation of the actual picking operation environment,it is found that the warehouse picking efficiency is low,and the order structure on site is relatively complex,the daily picking of goods overlap is low,and the overall planning of the warehouse lacks strict category inventory management.To address these issues,this paper designs and implements a new picking system based on the goods-to-man picking model to facilitate the execution and management of corporate warehousing operations.First,this paper analyzes the requirements of the main business in the process of warehousing operations;second,based on the requirements analysis of the system,the database structure and specific functions of each functional module of the system are designed and implemented.In addition,in view of the low efficiency of on-site picking,the complexity of orders and the low degree of commodity overlap,an order combination processing strategy is established and the order processing activities of the outbound module are optimized.The optimization mainly focuses on the low degree of overlap of commodities within orders and the current situation of mixed placement of various types of commodities,proposes a commodity relevance algorithm and an order relevance algorithm with commodity relevance as an important parameter,and implements an order combination function based on the idea of hierarchical clustering to combine orders with high relevance for picking,reducing the number of shelf moves during picking operations,and thus improving the efficiency of commodities leaving the warehouse.Finally,the main business functions and picking performance of the system were tested based on commodity and order data from the actual warehouse environment,proving that the system can already meet the business requirements of warehouse operations.Compared with other picking systems,this system is compatible with warehouse management and warehouse execution,realizes three operation modes of inbound,outbound,and inventory,and unifies the three different operation modes into a state-machine-based task processing engine that provides a common processing mechanism for task management,execution,and exception handling and recovery,enhancing maintainability and scalability.Moreover,because of the complex order structure,low commodity overlap,and lack of management in the early stage of this research object picking,it is impossible to improve picking efficiency from the perspective of order content alone,so this paper combines order correlation and existing distribution of inventory,designs and implements an order combination strategy based on hierarchical clustering,and optimizes the efficiency of outbound picking operations.