Home Internet of Things Aerospace Apparel Energy Defense Health Care Logistics Manufacturing Retail

Amazon Aims to Revolutionize Brick-and-Mortar Shopping

The newly introduced Amazon Go concept store appears to use RFID, video, Bluetooth and artificial intelligence to enable shoppers to simply walk into a store, pick up items and walk back out again.
By Mark Roberti
Dec 11, 2016

Last week, Amazon made news with the introduction of Amazon Go, a concept store that completely eliminates the point of sale. The store is currently open only to Amazon employees (it's in a Seattle office tower housing Amazon personnel), but the company plans to open the site to the public next month.

Customers scan a QR code on their smartphone as then enter the store, enabling the store to identify them. As they walk through the aisles picking up items, those products are automatically associated with their account, presumably through a combination of video analytics and Bluetooth. A video released by Amazon claims it has "woven the most advanced machine learning, computer vision and [artificial intelligence] into the very fabric of a store, so you never have to wait in line."

The video does not mention radio frequency identification, though an Amazon patent filing details a system that uses RFID to detect when a shopper removes an item from a shelf and then links that data to a handheld device.

Amazon has so far declined to speak to our reporters about the technology behind Amazon Go, though it has told us that it may be more prepared to share details in January, after the store is open to the public. My guess is that the company might be blending a variety of technologies to achieve its goal.

There are systems that use imperceptible pulses of light reflected off objects to enable them to be identified. These could be used instead of RFID to detect which items customers pick up. Bluetooth could be used to confirm who is standing in front of a particular shelf, while video analytics and artificial intelligence could then determine which of two or three shoppers picked up the item.

If RFID is not used to determine what was picked up, that could make the system less expensive. But it would also make it less useful, since the company would not be able to count inventory quickly and effectively, though video analytics could help identify items that are out of stock in a convenience store setting. (It would not be possible to do this in an apparel store, since video cannot indicate the difference between small, medium and large shirts folded on a shelf.)

Regardless of what technology is being used, the Amazon Go system could represent a game change for brick-and-mortar retailers—and for the RFID industry. Remember, Amazon was the company that got the world used to the idea that you could access your computer, order something and have it arrive a few days later, without having to leave home. That doesn't seem revolutionary today, but it certainly was when Amazon first introduced the idea.

If the company now gets people used to the idea that they should be able to walk into a store, find what they want every time and walk out without having to stand on any lines, every retailer will need to adapt to changing customer expectations. RFID will surely be necessary to make such systems a reality in a variety of retail formats.

Mark Roberti is the founder and editor of RFID Journal. If you would like to comment on this article, click on the link below. To read more of Mark's opinions, visit the RFID Journal Blog, the Editor's Note archive or RFID Connect.

  • Previous Page
  • 1
  • Next Page

Login and post your comment!

Not a member?

Signup for an account now to access all of the features of RFIDJournal.com!

PREMIUM CONTENT
Case Studies Features Best Practices How-Tos
RFID JOURNAL EVENTS
Live Events Virtual Events Webinars
ASK THE EXPERTS
Simply enter a question for our experts.
JOIN THE CONVERSATION ON TWITTER
Loading
RFID Journal LIVE! RFID in Health Care LIVE! LatAm LIVE! Brasil LIVE! Europe RFID Connect Virtual Events RFID Journal Awards Webinars Presentations
© Copyright 2002-2017 RFID Journal LLC.
Powered By: Haycco