The best way to combine humans and robots for ideal results in warehouse work is for employees to circulate between working with various co-bot partners, not with a single dedicated robot, according to new research from Erasmus University in Rotterdam, the Netherlands.
Under many real-world conditions, warehouse workers achieve higher productivity when they dynamically switch among multiple autonomous mobile robots rather than work exclusively with a single robot.
That was the conclusion of a study titled “Picking the Best Bot: Collaboration Strategies for Humans and Bots in Order Pick Systems with Traveling Salesman Problem Routing.” The research applied analytical modeling and simulation to evaluate more than 12,000 warehouse scenarios. It was recently published in Transportation Science, a journal of the Institute for Operations Research and the Management Sciences (INFORMS).
According to researchers, the findings challenge a common assumption that fixed human-robot pairings are the most efficient approach. Instead, they found that a flexible “swarm” policy, in which workers collaborate with different robots throughout a shift, often outperforms more rigid one-to-one assignment strategies. And as robots become faster and more plentiful, the advantages of the swarm approach grow even stronger.
“This is not simply a question of adding more robots,” said the study’s lead author, Mahdi Ghorashi Khalilabadi of Rotterdam School of Management, Erasmus University. “Our findings show that the way humans and robots are organized can have a major impact on throughput. When robots are faster or more plentiful than human pickers, allowing flexible collaboration can significantly improve performance.”
The study is significant because the rapid growth of e-commerce has increased the demand for efficient order picking systems in large warehouses. Many facilities have replied by deploying autonomous mobile robots (AMRs) to assist human pickers and improve throughput performance. This study focuses on two popular policies: the swarm policy, in which pickers switch between AMRs while picking, and the system-directed policy, in which a picker completes an order with a single AMR.
The results indicate that the swarm policy generally provides higher throughput than the system-directed policy, with gains increasing in the AMR-to-picker count and speed ratios. The system-directed policy is more effective when AMR and picker speeds are similar, the orders are large, and there is a limited number of AMRs.
