Finding the Best Ways for Humans and Robots to Work Together Requires "Swarm" Thinking, New Research Finds

Study shows flexible human-robot collaboration can significantly improve warehouse productivity

BALTIMORE, June 3, 2026 — If the future of warehouse work belongs to humans and robots working side by side, a key question remains: What is the most effective way for them to collaborate?

New research published in Transportation Science, a journal of INFORMS, suggests the answer may be more flexible than many warehouse operators expect. The study, "Picking the Best Bot: Collaboration Strategies for Humans and Bots in Order Pick Systems with Traveling Salesman Problem Routing," found that 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.

The findings challenge a common assumption that fixed human-robot pairings are the most efficient approach.

Instead, researchers 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. As robots become faster and more plentiful, the advantages of the swarm approach grow even stronger.

The study arrives as warehouses around the world continue investing heavily in automation to meet growing demand while addressing labor shortages and operational pressures. While autonomous mobile robots have already improved warehouse efficiency, relatively little research has examined how the structure of human-robot collaboration itself affects performance.

"This is not simply a question of adding more robots," said Mahdi Ghorashi Khalilabadi of Rotterdam School of Management, Erasmus University and lead author of the study. "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 researchers analyzed how workers and robots performed across a wide range of operating conditions commonly found in warehouses, including differences in travel times, order sizes and facility layouts.

The study focused on two of the most common human-robot collaboration approaches. Under a swarm policy, workers interact with multiple robots while completing orders. Under a system-directed policy, workers complete an entire order with a single robot before moving to the next task.

Using analytical modeling and simulation, the research team evaluated more than 12,000 warehouse scenarios.

The results were clear.

The swarm policy generally delivered higher order throughput than the system-directed approach. Performance gains increased as the ratio of robots to workers grew and as robots gained speed advantages over human pickers.

The researchers also found important exceptions. Fixed one-to-one pairings performed best when robots and workers moved at similar speeds, orders were relatively large and the number of available robots was limited.

In other words, there is no universally optimal strategy.

"Our modeling approach helps managers identify when the benefits of flexibility outweigh the costs of coordination," said Debjit Roy of the Indian Institute of Management Ahmedabad and co-author of the study. "It provides practical guidance for organizations designing new robotic systems or expanding existing operations."

Beyond comparing collaboration policies, the research offers insights into warehouse design, fleet sizing and automation investments. For some organizations, particularly smaller operations, the number of robots required to realize the full benefits of swarming may not be economically feasible.

For managers deciding how to deploy automation, the study suggests that success depends not only on the technology itself but also on how people and machines are integrated into a larger operational system.

"This research shows there is no one-size-fits-all solution," said René de Koster of Rotterdam School of Management and co-author of the study. "The best collaboration strategy depends on the operating environment. Our framework helps decision-makers identify the approach that best fits their specific warehouse."

As organizations increasingly turn to automation, the study highlights a broader lesson familiar to operations researchers: performance depends not only on the capabilities of individual technologies but also on how complex systems are designed.

Link to full PDF of study here.

About INFORMS and Transportation Science

INFORMS is the world’s largest association for professionals and students in operations research, AI, analytics, data science and related disciplines, serving as a global authority in advancing cutting-edge practices and fostering an interdisciplinary community of innovation.

Transportation Science is a premier peer-reviewed scholarly journal focused on research about all modes of transportation, present and prospective, and looks at planning and design issues and the related economic, operational and social concerns. It is published by INFORMS, the leading international association for data and decision science professionals. More information is available at www.informs.org or @informs.

###

Media Contact

Rebecca Seel

Public Affairs Specialist, INFORMS

[email protected]

(443) 757-3578