Human-Centric Order Batching and Allocation When Picking With Robot Teammates in the Warehouse

More e-commerce warehouses are deploying automated guided vehicles (AGVs) to assist human pickers in order picking, yet this collaboration presents unique challenges. Human pickers often struggle to keep pace with their robotic counterparts, leading to significant physical and psychological strain. Frequent rack visits can result in non-ergonomic injuries, while unbalanced task allocations may induce mental stress, ultimately decreasing operational efficiency and worker well-being. Unlike previous studies that primarily focus on system-level efficiency, this study adopts a human-centric perspective to address these issues. We develop integrated optimization models for order batching and batch allocation aimed at minimizing human rack visits and promoting balanced workload distribution among pickers. Efficient heuristic algorithms are proposed to solve these models, and comprehensive computational experiments evaluate their performance. The results show that the optimized order batching strategy can reduce rack visits by up to 11.86%, mitigating ergonomic risks, while the optimized order batch allocation strategy significantly improves workload balance with negligible computational cost, contributing to fairer task distribution and reduced psychological stress on human pickers. The robustness and effectiveness of the proposed algorithms are rigorously validated, making a significant theoretical and practical contribution to enhancing human-machine collaboration in order picking and promoting human-centric warehouse management in the Industry 5.0 era.