Study: RFID Scan Accuracy Drops As Number Of Items Increases
Written by Eric AthasAugust 7th, 2008
The more RFID-tagged items retailers place on Z-bars or shelves, or in boxes, the lower the read rate will be when those items are scanned, according to a study out of the University of Arkansas. By setting up store-like scenarios in a university lab, researchers conducted three feasibility test scenarios of RFID tagged apparel and shoes.
What they found was that when an associate scanned items, the number of items scanned affected the read rate. For example, researchers scanned RFID items hung on a rounder. When there were 97 items, they found a 99.38 percent read rate. Compare that to the 89.89 percent accuracy of 180 items scanned.
2 Comments | Read Study: RFID Scan Accuracy Drops As Number Of Items Increases
Leave a Reply
Readers, specifically those who want to comment on a story:
Our Comment SPAM system is getting very aggressive these days and has been blocking legitimate comments. If you post a comment and don't see it appear within 2 hours or so, can you please send a heads-up to customer-service@storefrontbacktalk.com? Ideally, please include the time you posted the comment. That will allow us to try and hunt for it. Thanks! P.S. We're working on fixing the system, but we don't want to lose any valuable comments in the meantime.
Our Comment SPAM system is getting very aggressive these days and has been blocking legitimate comments. If you post a comment and don't see it appear within 2 hours or so, can you please send a heads-up to customer-service@storefrontbacktalk.com? Ideally, please include the time you posted the comment. That will allow us to try and hunt for it. Thanks! P.S. We're working on fixing the system, but we don't want to lose any valuable comments in the meantime.
-Christine
August 14th, 2008 at 10:08 am
One element of this application is the reading of tags on clothing hanging on a Z-bar. Using a mobile RFID reader, the user is asked to aim the reader at the clothes in various orientations passing it back and forth and around until all the readings were recorded.
What bothered me about this procedure is the question as to when, and how the user determines that all the tags are read.
On the last page of this study I found the following sentence:
“In the course of evaluating read rates to determine how a test should proceed, each tester often had to know how many tags they were attempting to read during a test. In practice, however, this might often not be the case.”
I believe the last sentence of this quotation to be incorrect. It is safe to say that in practice this is never the case.
I believe the read rates would be much lower in any practical environment where the user does not know the number of tags involved, and has to use judgment to determine when to stop. Furthermore, the results would be highly variable from user to user depending on training and commitment to the task.
The study can be downloaded from here:
http://itri.uark.edu/91.asp?code=&article=ITRI-WP112-0608
August 15th, 2008 at 9:36 am
In my experience with Item Level RFID tests, I have found that today’s RFID HandHelds do are not ready for doing cycle counts of item level tagged product. Also, RFID HandHelds are much more sensitive to the quality of the RFID tag.