RFID and Automatic Identification Focus: Our Weekly Feature Article on Topics of Interest to those Using or Considering RFID or other Auto ID Technologies  
  - August 25, 2009 -  

RFID News: Second Item-Level Retail Apparel Tagging Study Again Finds Substantial Improvements in Inventory Accuracy

pdf of this article

Will Item-Level Tagging in this Space be the Catalyst for Case and Pallet RFID?

  SCDigest Editorial Staff  


SCDigest Says:

In the control store, the inventory audits found substantial declines in PI accuracy during the course of the 13-week test period.

Click Here to See Reader Feedback

The Information Technology Research Institute at the University of Arkansas, under the leadership of Dr. Bill Hardgrave, has just released another research paper on the impact of item-level RFID tagging in the apparel retail sector. This time, the study was based on work at Bloomingdale’s department store, a division of Macy’s. This is a follow up, in a sense, to a very similar study the Institute performed earlier this year at Dillard’s. (See New University of Arkansas Study on Item-Level Apparel Tagging Interesting, but Cause-Effect Relationships Not Clear.)


As in the previous study, the researchers first created a baseline by comparing the inventory that a store thinks it has, via its perpetual inventory system (PI), to the actual inventory on-hand, which was determined by an outside auditing firm.


For any individual SKU, the actual inventory can be “perfect” – equal to what the PI says, an understock (there is less actual inventory than the PI indicates), or an overstock (there is more inventory that in the PI). Under and overstocks result from a myriad of causes, including shrinkage, POS errors, in-store cycle counting errors, and mis-shipments from the retail DC or a vendor. These errors can have a significant cost in terms of lost sales (understocks) and excess inventories (overstocks).


In the control store, the inventory audits found substantial declines PI accuracy during the course of the 13-week test period, as shown in the graphic below.





(RFID and Automatic Identification Article - Continued Below)





One store was outfitted with RFID readers at several points (receiving, entrance to store from back room, POS). Another control store relied on barcode scanning and/or manual counts and methods.


During the test, the PI in the RFID test store was not actually changed. Rather, the PI was simulated, with the RFID read/transaction data being posted to that simulated system, not the live system at Bloomingdale’s.

According to the study, the level of “perfect” inventory accuracy in the RFID store increased by 27%, as measured in the simulated system versus the actual PI. The preponderance (21%) of that improvement came from reducing the number of understocks.


Unfortunately, it appears that Bloomingdale’s would not allow actual numbers from the study to be published in the public report. Therefore, it is difficult to get a real feel for what this 27% improvement really means, and what the economic value really would have been, although intuitively, it seems likely that it would not be trivial.


As with the previous study, however, this effort was able to quantify the productivity improvements from RFID-based cycle counting, which was found to be some 96% faster than with barcoding. These efficiency gains would enable a retailer to perform cycle counts much more often (and much more accurately), further improving PI accuracy.


The report notes that for apparel, at least, the presumed progression may be reversed – item-level tagging may actually drive case and pallet level tagging, or even serve as a proxy for tagging at that level, rather than the general expected paradigm that first pallets would be tagged, then cases, then units.


The full report is available here: Bloomingdale’s Item-Level RFID Test.


What is your reaction to this study? Do percentage numbers without any real detail help your understanding? Do you expect to see widespread item level tagging in apparel retail any time soon? Let us know your thoughts at the Feedback button below.

SCDigest is Twittering!

  Follow us now at https://twitter.com/scdigest

Send an Email