No Free Lunches With RFID

I frequently find myself at meetings—both academic and industrial—where discussions center on the availability of vast amounts of “free” information that simply needs to be mined or processed appropriately to reveal great business “nuggets.”

No doubt we can access huge amounts of data with increasing ease, but that free information is generally not the information we want. And even if the information we need were free in monetary terms, it wouldn’t be free in terms of energy. Information and communication technologies are some of the planet’s great carbon dioxide producers, forecast to surpass air travel in that regard in several years.

We need to accept the fact that there’s a cost for any information we gather, so we must understand that data’s value before we commit to collecting it. In this respect, RFID is no different from any other information source.

In the RFID gold rush early this century, mandates often outweighed the need for any cost-value analysis. Then, as the situation inevitably tightened, RFID’s value dropped to a cost-reduction calculation—either labor force or inventory reduction—for which simple cost-benefit analyses could determine payback on investment. Now, organizations are using RFID to address more complex benefits. For example:

• How can we achieve more certainty about the arrival of baggage at an aircraft, and enable determination of optimal take-off time?

• Can a real-time location system increase confidence in the location of tools in a factory and reduce production delays?

• Is it possible to reduce the risk of out-of-stocks in a supermarket with RFID-driven reordering?

• How can we decrease the chance of using the wrong replacement part?
Issues of certainty, confidence, risk and chance all relate to the probability of people making correct decisions and taking proper actions. Understanding and quantifying the underlying value (and cost) of RFID-derived information has been central to the past decade’s work at the Cambridge Auto-ID Lab, where we have developed tools and methods to address the value-of-information (VOI) questions associated with RFID use in manufacturing, logistics, retail, maintenance, asset management and recycling.

For example, “Would the cost of acquiring more accurate on-shelf availability information be less than the cost of having out-of-stock items, and losing potential sales and customers?”

This philosophy has been carried over into the lab’s spinout company, RedBite Solutions, where establishing RFID value for the end user is paramount. The lab is expanding its research to examine VOI issues in broader distributed sensing and data-collection environments, in areas such as construction and airport operations, where RFID is just one of many data sources that must be considered.

While there is no “free lunch” in information capture and processing, we want to make sure the user gets a satisfying, low-cost/good-value meal!

Duncan McFarlane is director of the Cambridge Auto-ID Lab.