Optimizing routes for order picking in warehouses.

Abstract

The present study investigates the application of mathematical optimization in the context of warehouse logistics, with a particular focus on the Split Delivery Vehicle Routing Problem (SDVRP) encountered by the e-commerce company, Ellos Group. The SDVRP requires a solution that partitions demand between forklifts while satisfying routing and computational constraints in a warehouse setting where forklifts are used to travel and collect items from shelves, and items may vary in volume. Since exact methods are time consuming, we propose a heuristic to identify good and feasible solutions for our optimization problem. Computational experiments based on practical (industrial) benchmark instances show that the proposed algorithm outperforms the existing routing policy method employed by Ellos in terms of solution quality. The results of this study suggest that the proposed algorithm has practical implications for improving the efficiency of warehouse operations and reducing transportation costs. The study contributes to the field of mathematical optimization by providing a practical solution to a complex real-world problem in the context of warehouse logistics. The proposed heuristic’s ability to solve the SDVRP with demand variability and routing constraints, along with constraints on computational time, demonstrates its efficacy in real-world settings. This work highlights the importance of mathematical optimization in modern industries and underlines the potential of heuristic methods in addressing complex real-world problems.