An independent study, commissioned by Wal-Mart and conducted by the RFID Research Center at the University of Arkansas, found a 16 percent reduction in out-of-stocks, as well as a reduction in excess inventory in RFID-enabled pilot stores.

In an industry that's hungry for good news, this seems like great news and shows a real benefit for manufacturers who are applying EPC labels.  However, upon closer reading, it may be more a matter of the process changes associated with RFID that should be credited with much of the improvement.

One would expect lower labor costs or more efficient utilization of labor because employees don't have to pick up a scanner to read a bar code (although scanners on conveyors can read a 14-digit case code automatically).  But labor wasn't included in the study's report.  The report examined stock levels and replenishment rates.

The study concluded that, overall, RFID-enabled stores were an astonishing 63 percent more effective in replenishing out-of-stocks than the control (bar code) stores.  Moreover, items bearing electronic product codes (EPCs) that could be read by RFID readers were replenished three times faster than comparable items marked with bar codes.  Finally, RFID-enabled Wal-Marts also were able to lower  stock levels and reduce manual orders by around 10 percent.

According to the news release, Bill Hardgrave, director of the RFID Research Center at the University of Arkansas and executive director of the Information Technology Research Institute stated, "Our analysis consistently found -- throughout the test period -- that the RFID-enabled pilot stores statistically outperformed the control stores without RFID technology in terms of providing improved on-shelf availability of items for customers."

Wal-Mart has always been known for employing technology intelligently.  Wal-Mart aggressively adopted bar codes to help manage inventory.  A well designed bar code receiving system can perform nearly as well, if not as well, as an RFID system in many situations.  Between scanning incoming shipments, scanning what goes onto the floor and scanned data from point-of-sale, the store should give a very good indication of what's on the shelf (some potential variation would exist because of shoplifting).

The question, therefore, is where did the benefits of RFID come in?  Or, to put it a different way, where did the existing systems fail that there was the opportunity to improve efficiency by 63 percent?

One explanation could be that the pilot stores were more efficient simply because they were pilot stores.  This is not an uncommon occurrence during pilot projects or when other types of tests are being conducted.

For example, in one industrial setting, the lighting level was adjusted to determine whether it would have an effect on productivity.  When the lighting level went up, so did productivity.  However, when it went down, productivity again went up.  Researchers found that productivity increased when workers felt the management either cared about their working conditions (were conducting tests) or, from a less positive perspective, when management was monitoring them.  (This second effect was only in short-term results.  Long-term management scrutiny often resulted in decreased productivity.)

Another explanation is that RFID systems were more closely integrated or monitored than existing bar code systems.  Again, from a less positive perspective, it's possible that bar code systems weren't being utilized efficiently or consistently.

At this point, a 63 percent increase in efficiency over bar code inventory seems a bit too good to be true, even in an industry where significant gains are the norm when old, inefficient systems are replaced by new, more efficient ones.