Human Aspects in Collaborative Order Picking – Letting Robotic Agents Learn About Human Discomfort

Human aspects in collaboration of humans and robots, as common warehousing, are considered increasingly important objectives operations management. This work aims to let robots learn about human discomfort collaborative order picking robotic mobile fulfillment systems. To this end, a multi-agent reinforcement (MARL) approach that considers next traditional performance the reward function agents is developed. As first step, we assume human-oriented assignment problem which assign orders workers at stations. The results show among four evaluated policies, only proposed MARL policy effectively discomfort. While may need be refined obtain near-optimal solutions for trade-off between efficiency objectives, it also shows practicable pathway related problems human-robot collaboration, inside outside warehousing.