Amazon has this week revealed details about its ‘Package Decision Engine’, an artificial intelligence (AI) model that helps it use less packaging than it otherwise would in customer orders.
The e-commerce and tech giant said it uses the AI engine to determine the most efficient type of packaging for each item it learns about, ultimately helping reduce the number of cardboard boxes and envelopes, paper dunnage, tape, and paper bags used to send goods to customers.
Built on the Amazon Web Services cloud, the model predicts when a more durable product like a blanket doesn’t need protective packaging, or when potentially fragile items, including crockery, require a studier box.
Amazon said it uses a combination of “deep machine learning, natural language processing, and computer vision”, and the model is continuously learning about Amazon’s operations to help it produce less packaging.
According to the statement released on Tuesday (16 April), prior to the use of AI, staff used physical testing on individual products to determine how to optimise packaging.
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.
“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.”
How it works
- When an item first arrives at the Amazon fulfilment center, it is photographed in a computer vision tunnel that determines the product’s dimensions, spots defects, and captures multiple images of the product. This also allows the model to detect if there is a bag or box around an item, or detect the presence of exposed glass.
- The AI model also uses natural language processing and leverages text-based data from each item, such as the item’s name, description, price, and package dimensions. It also collects information in near-real time from customer feedback that is reported through Amazon’s online returns centre, product reviews, and other customer feedback channels.
- After compiling the information, the model produces a score that predicts the best packaging type to use. The packaging selection is remembered by the model and used to understand future packaging needs.
- Amazon said its scientists have trained the AI model by showing it millions of examples of products that had been successfully delivered in various types of packaging without damage. They also showed it products that have arrived damaged, along with the keywords and packaging types used in each scenario.
- The model has learned that certain keywords are important when supporting the move to less packaging. For example, a padded mailer with limited cushioning might not adequately protect an item with the words “grocery,” “screen,” or “stoneware” in the description, so the model would recommend a sturdier option, such as a box. The model also learned that keywords like “multipack,” “bag,” “shrink,” and “pack” were also associated with lower damage rates in the mailer, and indicated the product might already have protective packaging and not need additional protection.
The model is already widely used in Amazon’s UK, European, and North American fulfilment centers, with the company saying this week that components are being rolled out in additional locations across India, Australia, and Japan.
[image credit: Green Retail World]
