The world’s largest e-commerce platform Amazon uses artificial intelligence in its warehouses to check items for damage before shipping orders to customers.
Amazon hopes the new technology will reduce the number of damaged goods shipped out, speed up picking and packing, and play an important role in the company’s efforts to automate more of its operations.
Amazon estimates that only less than one in a thousand items are damaged, but the total is significant for the retailer, which handles around 8 billion packages a year.
In light of labor shortages, Amazon has been constantly working to automate its warehouses recently and has tried to outsource some of the most physically demanding and repetitive warehouse work to robots.
Amazon’s introduction of artificial intelligence is part of a broader push across the business world to bring more AI technology to logistics operations as companies try to manage complex supply chains while keeping goods moving quickly and reliably.
Amazon has implemented AI in two fulfillment centers so far and plans to roll out the system to ten more locations in North America and Europe. The company has found that artificial intelligence is three times more effective than a warehouse worker at detecting damage, said Christoph Schwerdtfeger, head of software development at Amazon.
AI checks the items during picking and packing. Items are selected for individual orders and placed in boxes that pass through a photo-taking station where they are inspected to confirm the correct products have been selected. This imaging station also assesses whether items are damaged. If damage is detected, the box is transferred to an employee who inspects the goods again. If everything is in order, the order is moved forward to be packed and shipped to the customer.
According to Schwerdtfeger, Amazon trained the AI using photos of undamaged items versus damaged items, teaching the technology the difference so it could recognize a product if it didn’t look perfect.
Source: The Wall Street Journal