Improving Warehouse Slotting Using Clustering and Genetic Algorithm

Abstract

The optimization of warehouse operations, particularly order picking, is crucial for reducing operating costs and enhancing ergonomics to prevent work-related injuries. This thesis addresses the challenge of optimizing slotting in manually operated warehouses by integrating clustering and genetic algorithms to improve order picking efficiency and ergonomics. Using K-Medoids clustering, products were grouped into clusters, which were then strategically placed in the warehouse through a genetic algorithm to minimize picking distance and improve ergonomic conditions. The study further refined slotting by optimizing the placement of products within clusters. The results demonstrate that this AI-driven approach outperforms random slotting schemes, significantly reducing picking distance and enhancing ergonomic safety. Moreover, clustering using AI methods produces more well-defined and evenly distributed clusters compared to traditional ABC analysis. The study highlights the importance of the clustering function’s logic in achieving optimal warehouse slotting and suggests that a well-designed AI-powered slotting system can lead to substantial operational improvements. A quantitative case study method was employed to test the algorithm, confirming its effectiveness in a real-world setting. This research contributes to the field of warehouse management by demonstrating the effective integration of AI in slotting optimization. The findings provide valuable insights that can be applied across various industries, paving the way for more intelligent and effective supply chain solutions.