How AI helps Amazon reduce packaging waste

Amazon has introduced an AI-powered solution known as the Package Decision Engine to enhance packaging efficiency for customer orders. This cutting-edge technology is aimed at ensuring that purchases are shipped in the most appropriate packaging, thereby minimising waste and improving delivery processes.

Built on Amazon Web Services, the Package Decision Engine employs a sophisticated multistep process to analyse each item. Initially, products are scanned in a computer vision tunnel at Amazon’s fulfilment centres to determine dimensions, detect defects, and capture various angles, including identifying whether an item is already boxed or contains exposed glass.

Moreover, the AI model utilises natural language processing to extract text data such as product names, descriptions, prices, and packaging dimensions. It also incorporates near-real-time customer feedback from Amazon’s Online Returns Center, reviews, and other communication channels.

Combining deep machine learning, natural language processing, and computer vision, this multimodal AI model can predict when items, like blankets, require less protective packaging or when fragile goods, such as dinner plates, necessitate sturdier boxes.

Kayla Fenton, senior manager of technology products with Amazon’s Packaging Innovation team, said: We wanted the ability to quickly identify the most efficient packaging option for each item, while also predicting how safely each product would ship, which includes research scientists and technical program managers. The use of AI through the Package Decision Engine has allowed us to advance our packaging efficiency work at scale quickly, and it has worked so well that we’re implementing this technology across Amazon’s broader global footprint.

Before implementing AI, Amazon relied on manual testing to optimise packaging, a method that was reportedly not scalable.