0
0
June 5, 2023 by Pleuni.
With artificial intelligence, Amazon is screening items for damage before they are shipped. This will reduce the number of damaged items sent out and speed up the picking and packing process.
During the order picking process, warehouse workers check items for signs of wear and tear. Since most items are in fine condition, checking for damage can take a long time, according to Amazon Robotics director Jeremy Wyatt. You are looking for something rare, and it is not your primary job, so it is cognitively demanding.
The number of damaged items is less than one in 1,000
It handles about 8 billion packages a year, so the total number of damaged items is significant. However, less than one in 1,000 items is damaged. The ecommerce giant has recently automated its warehouses. The company recently added robots to its warehouses in the United States as it struggles to hire workers.
In the US and Europe, Amazon is rolling out the system in 12 locations.
According to Amazon, the AI has proven three times more effective at identifying damage than warehouse workers at two of its fulfillment centers. It is now rolling out the system to 10 other sites in North America and Europe.
Picking involves checking items
This AI was trained by comparing photos of undamaged with damaged items, in order to be able to flag a product when it does not look perfect during the picking and packing process.
It is moved to a worker for inspection if something is broken.
The goods are picked and placed into bins for individual orders. They move through an imaging station, which checks to see if the right products were selected. Now, the imaging station checks to see if anything is damaged. In the event that something breaks, the bin is moved to a worker.
