Amazon develops algorithm to improve cooperation between robots and humans

Amazon has identified a financially profitable way for robots and humans to coexist, while saving online businesses half a billion dollars a year. Using robots to carry inventory shelves to associates to pick customer orders reduces the robot’s travel distance and reduces the company’s storage space.

A common fear is artificial intelligence (AI) and robotics will put Americans out of jobs, but it may not happen—new research on INFORMS Journal of Applied Analytics is using operations research (OR) science to enable humans and robots to work together more effectively.

Study “Robot picking algorithms in Amazon’s fulfillment centers enable humans and robots to work together efficiently,” shows work being done to redesign robotic picking algorithm used in Amazon Robotics (AR) fulfillment centers (FCs).

“Instead of associates going to product shelves to select shipments for customer orders or stocking new inventory, robots bring inventory shelves to associates,” said Amazon’s Russell Allgor. at the workstation to pick or queue,” said Russell Allgor of Amazon.

The order picking algorithm decides which specific inventory units should be picked on which buckets to fulfill shipments to customer orders. The algorithm reduced the distance traveled by groups by 62% with no negative impact on performance. This is now implemented in all Amazon Robotics FCs.

“This improvement reduced the number of drives required in AR FC by 31%, saving half a billion dollars,” Allgor continued.

Allgor, along with co-authors Tolga Cezik and Daniel Chen, both of Amazon, say that by implementing AR FC as the standard for new fulfillment centers, Amazon has also reduced storage capacity by about 29%. compared with the non-AR FCs. This means more efficient use of space and a more productive human workforce.