Home Amazon Using AI To Detect, Divert Defective Product
June 4, 2024

Amazon Using AI To Detect, Divert Defective Product

Posted In: Retail Articles

Amazon is rolling out an AI-enabled function, dubbed Private Investigator, that flags damaged products, improves the customer experience, prevents waste and enables reselling.

Across North America, Amazon fulfillment centers pass through imaging tunnels for examination as part of the company’s Project P.I. By using a combination of generative AI and computer vision technologies. The company noted that Project P.I. uncovers defects, including product damage, colors different than those ordered and incorrect sizes before items reach customers. In addition, Project P.I. helps identify the root cause of issues, enabling preventative measures upstream that prevent reoccurrence. At the sites where the system is in operation, Private Investigator has proven adept at sorting through the millions of items that pass through the tunnels each month and accurately identifying merchandise issues, according to Amazon.

Amazon associates who review the items Project P.I. flags then decide whether a product is eligible to be resold at a discounted price as part of Amazon’s Second Chance site, considered for donation or diverted to another use. The model acts as an enhancement to manual inspections at fulfillment centers where it’s in use and, so, is another tool to help prevent product returns, which diminishes wasteful consequences in transportation and related operations. Amazon pointed out that it plans to expand Project P.I. to additional sites throughout 2024.

Project P.I. is an Amazon AI application developed to enhance customer satisfaction and contribute to its efforts to address climate change. It is one of several related initiatives that also includes the Package Decision Engine, an AI model that helps ensure customer purchases arrive in the most efficient packaging available while preventing damage and making deliveries easier.

In parallel, Amazon is leveraging a generative AI system that uses a Multi-Modal Large Language Model to investigate the root cause of negative customer experiences, the company maintained. When a customer informs the company of a defect it failed to identify, Amazon applies the input to understand the cause and continuously improves its systems in part by analyzing Project P.I. image output. Amazon is getting ready to use the same technology to help its selling partners access defect data and prevent future errors, it added. 

“We want to get the experience right for customers every time they shop in our store,” said Dharmesh Mehta, Amazon vice president of worldwide selling partner services, in discussing Project PI. “By leveraging AI and product imaging within our operations facilities, we are able to efficiently detect potentially damaged products and address more of those issues before they ever reach a customer, which is a win for the customer, our selling partners, and the environment.”

Share Now!

Related Posts: