RFID: The Good, The Bad, And The Ugly

If you've read any trade or even mainstream press in the last few months, you'll have noticed the growing momentum--some would say hype--behind RFID.

Pulled by the desire to reduce costs and increase information in the supply chain (some estimate that billions of dollars of excess inventory can be eliminated with better information) and pushed by mandates from the likes of Wal-Mart and the U.S. Army, radio-frequency identification tag usage looks set to explode in the coming months. The multibillion tag market that these two customers alone account for will create dramatic economies of scale, pulling more and more potential applications into the realm of economic feasibility, leading to a "virtuous circle" of falling prices and rising adoption.

Unfortunately, there's a shortage in some corners of critical thinking about what RFID is good for and not so good for, something that always seems to happen with radical new technologies. As with other process-busting innovations of the past, such as the dot-com revolution and the telephone (one early business model for the telephone was to deliver background music to restaurants), the ideas that are floated can be divided into the good, the bad, and the ugly.

Among the good ideas are the obvious supply-chain improvements that come from being able to better track goods every time they pass a reader-enabled checkpoint. More innovative ideas come from combining simple identity tags with slightly more complex active sensors that can measure their environment. For example, the U.S. Army is doing trials on a system in which food rations for the troops, called Meals Ready to Eat, in military jargon, are tagged with cheap identifying tags, while the crates they are shipped in carry active tags that measure the ambient temperature. This is vital information for the Army because the shelf life of MREs drops dramatically when they are stored in, for example, the heat of a desert camp in Iraq. Similar technology has applications in the civilian world, for example, to monitor food in refrigerated railroad cars or to ensure that medicines are maintained at appropriate temperatures throughout the delivery chain.

Other good ideas outside the supply chain include tracing objects through their life cycle. One application is baggage handling: Airports and airlines are interested in using RFID tags to more efficiently and accurately handle passengers' luggage. In the past, automated baggage handling has often been limited by the need to scan bar codes. In a high-tech system such as the one at Denver International Airport, bags are scanned from all directions by an array of scanners looking for the bar code, because it's so hard to control the position and orientation of each bag's tag. With even the best systems, as many as 10% of bags can't be read automatically and have to be diverted to be scanned manually. RFID tags would largely eliminate this problem--in fact, bags wouldn't even need to pause for reading, but could whisk past the reader at full speed. In addition, RFID tags can be read repeatedly, verifying a bag's identity and destination each time it is routed onto a conveyor or cart, reducing the risks of a bag accidentally being sent to the wrong destination because of handling mishaps.

However, you don't have to get too far beyond these good ideas before you find yourselves in the realm of the bad (or at least the badly thought-through). One abandoned retail RFID trial involved tags on razor blades combined with cameras that would automatically photograph each person who took a packet of blades off a "smart shelf". This would supposedly reduce shoplifting--razor blades are apparently a popular item--by, for example, identifying individuals who are seen to take a packet of blades but never recorded as paying for those blades at the checkout. The potential problems with this idea are myriad, not least the risk of accusing an innocent shopper who simply changes their mind about the blades and leaves them on another shelf in the store. Furthermore, smart shoplifters would quickly figure out that the easiest way to bypass the system is to wait for somebody else to get photographed taking the item off the shelf, then steal it from their basket!

And then there are the ugly ideas. By far the ugliest is one that was raised recently by a vendor that, we hope, is merely crying for attention rather than seriously proposing this. Their idea is to have consumers submit to surgery (under local anesthetic) to implant RFID chips under the skin that would be used for electronic payments. The supposed rationale for this implant is that, unlike cards or fobs, you can't lose or forget it. Contrast that with the downside--that you can't leave it behind even if you want to, not to mention the costs, the risks, and privacy implications (and then there are the difficulties of distributing upgrades!). The privacy advocates' outcry about a tag that really can trace an individual whenever they're near a reader is easy to imagine.

If ugly ideas like this were merely forgettable, it would hardly be worth mentioning. Unfortunately, there's already a body of misinformation about RFID that's feeding privacy and confidentiality concerns and obscuring rational debate about the role, value, and desirability of the technology. There are plenty of good ideas for the technology; we can even live with the bad, as the marketplace will, for the most part, sort them out. But we could really do without the ugly.

Carl Zetie is an analyst with Forrester Research.