Why does Amazon want to kill barcodes?

Today, in Amazon fulfillment centers, employees must identify products at various points by manually scanning barcodes. As the e-commerce giant explains, the employee must take the item, look for its barcode, then scan it.

The company is not satisfied with the performance achieved using this system, which is based on technology that is almost 50 years old. “This process is repeated millions of times across a huge catalog of items of varying shapes and sizes, and it cannot be easily automated,” Amazon explains.

Indeed, for the moment, there are no robots powerful enough to automate this task. Also, Amazon’s goal is now to say goodbye to the barcode, using a new method of identifying items based on AI.

Amazon wants to replace the barcode with the MMID

In order to get rid of the barcode, Amazon researchers are working on a new method of product identification, called multimodal identification or MMID. In essence, it involves identifying an item from an image, extracting information about its appearance and dimensions. And while the system isn’t perfect yet, Amazon researchers believe it may one day stop using barcodes in warehouses.

“This vision, to use MMID throughout the execution process, to accelerate and enable robotic automation, is going to be achieved”says Nontas Antonakos, one of the initiators of this project at Amazon. “And when it does, it will be another step in our journey to get packages to customers faster and more accurately.”

It’s not the data that’s missing

To begin, Amazon researchers photographed items moving on the conveyors. Then, these images were converted into vectors, sequences of numbers. Next, Amazon created machine learning (AI) algorithms for identification.

In the first experiments, Amazon researchers obtained a match rate (thus, correct identifications) of 75% to 80%. And today, they would be at 99%.
This high rate is explained by the fact that the algorithms do not search for the item to be identified through the entire Amazon catalog. An impossible task, according to the company.

“Each item comes from a particular package, and each package contains a few dozen products. So the algorithm only has to match an item to the contents of a single package”specifies the e-commerce giant.

In any case, it is not the data to perfect this system that is lacking, since Amazon only has to position a camera above its conveyors. This provides images of scrolling items, to train the AI.

Potential and limits

For Amazon, this technology is promising. And it could pave the way for more automation in its warehouses. Moreover, in Hamburg and Barcelona, ​​this artificial intelligence is already proving its worth. On the conveyor belts, it detects errors: when an item does not correspond to what is indicated in the inventory.

But the e-commerce giant also admits that its technology has limits, for now. For example, this is tested on conveyor belts because the lighting as well as the speed at which items move are relatively constant. At other locations, identification might be more complicated.

We would love to give thanks to the author of this short article for this awesome web content

Why does Amazon want to kill barcodes?

We have our social media profiles here , as well as other related pages herehttps://www.ai-magazine.com/related-pages/