A Study on Storage allocation problem based on clustering algorithms for the improvement of warehouse efficiency

The operation of warehouses has long been a focus of industry research. Faced with rapidly growing business needs, improving storage efficiency, and reducing customer response times have become crucial issues for improving the operational efficiency of a warehouse. Given a fixed area of space, optimizing the storage strategy can reduce the cost of goods handling, improve the efficiency of storage and delivery, accelerate the overall operational efficiency of the warehouse, and reduce logistical costs. In this paper we study the improvement of a real-life company’s storage location strategy using cluster and association analysis. Two different clustering techniques namely pairwise comparison clustering and K-means clustering are used, and their performances are compared with the current random storage policy used by the company. Both clustering algorithms consider item association and classify items into groups based on how frequently they appear with each other in customer's orders. The next stage applies assignment techniques to locate the clustered group in each aisle so as to minimize the total number of aisle visits and ultimately picking distance. By emphasizing the item association, our model is suitable for orders with multiple items in the modern retailing sector. It also more effectively shortens the picking distance compared with random assignment storage method. In our case, Warehouse studied herein, both models prove more effective as it reduces over 35% and 25 % of the picking distances versus the current set-up. However, when compared with each other the K-means clustering method outperforms the pairwise comparison