Warehouse Order-Picking Digital Twin for Storage Assignment, Order Batching and Picker Routing

Order picking is the most expensive activity in a warehouse, and most of it is walking. This project maps two years of real order lines of a UK online retailer (UCI Online Retail II) onto a documented simulated 10-aisle, 1,000-slot warehouse and compares classical and data-driven policies for the three decisions that drive the walking distance: where each product is stored (random, ABC, turnover vs a gradient-boosting demand forecast and co-occurrence clustering), which orders are picked together (FCFS vs seed, k-medoids and a surrogate-guided local search) and the route (S-shape and largest gap vs the exact optimal tour). Slotting uses only past weeks; results are measured on the later weeks. A SimPy digital twin replays picking days with several pickers and aisle congestion, and a web application shows an animated warehouse view, pick waves and pick lists. A complete MATLAB version is included.

Category: Industrial Engineering

Industrial Engineering Project IE07 from ProTechSC Academy, with graphical abstract, video demonstration and project guidance.