Why does adoption of RFID always seem to lag behind forecasts? Experts say there are good reasons, including:
- Benefits aren’t always easily understood by potential users.
- Process changes take time, especially when people must be re-trained and machines must be re-programmed.
- Developing the infrastructure for RFID calls for major capital investment.
From the pages of Design News
|RFID: Beyond the Drive for Five|
|Charles J. Murray, Senior Technical Editor — 4/24/2006|
For almost a decade, the “nickel tag” has loomed unattainably on the horizon. Chief information officers have waited patiently for it. Experts have forecast a role for it in “trillions” of everyday products. The European Central Bank has even been rumored to be hatching a plan to weave the five-cent Radio Frequency Identification (RFID) tags into the fabric of its currency. There is, it seems, no limit to the reach of RFID once that five-cent goal is attained.
Still, vendors are showing there’s more to RFID than nickel tags.
“We’ve been talking about the mythical five-cent price point for years,” notes Mike Liard, RFID practice director for Venture Development Corp. “Is it possible? Yes. But it may not necessarily be the type of tag you’re looking for.”
Indeed, makers of RFID chips and so-called “inlays” (which include chip, antenna and substrate) know this, which is why most haven’t rushed to put nickel tags on the market. Instead, they’ve been content to cut prices at a steady rate of about 5 to 10 percent per year since 2000, while simultaneously improving the technology. As a result, users of the tags are already employing them in applications undreamed of a decade ago, despite their inability to reach the elusive nickel price point. At McCarran International Airport in Las Vegas, for example, “bag tags” with dual dipole antennae are attached to luggage to ensure that RFID readers in the handling system can communicate with all bags, no matter their orientation on conveyor belts. Such dual-antenna tags haven’t reached rock-bottom price points, but at roughly 20 cents apiece, they offer capabilities nickel tags couldn’t match today. Similarly, retailers have begun using tags with specialized antennae to enable garments buried in stacks to successfully “talk” to RFID readers. Again, cheaper tags would be unlikely to achieve such feats.
The bottom line is that while RFID vendors have been lowering their prices and improving their technology, they’ve been carving out new niches for themselves. Increasingly, RFID tags are being used on pallets, cartons, garments, luggage, DVD cases, pill bottles and library books. And in the future, experts foresee more use in low-cost everyday items, from lipstick cases to cereal boxes. While they won’t replace the bar code any time soon, they nevertheless offer non-line-of-sight capability, which means they can gather information of their whereabouts without being individually handled. As a result, they’re capable of deterring theft and counterfeiting.
“RFID is not labor intensive,” notes Sanjay Sarma, associate professor of mechanical engineering at Massachusetts Institute of Technology and research director for MIT’s Auto-ID Center. “It gives you information you can’t get with a bar code, unless you have an army of people scanning every product.”
“The state of momentum within the industry toward the five-cent mark is very healthy,” Sarma says. “The good news is that it has gone beyond research. It’s moved into development, and a lot of companies are looking to go to five cents.”
Whether or not five cents is their ultimate goal, makers of chips and inlays alike have targeted lower cost. Chip designer Impinj Inc., for example, is cutting costs through a novel semiconductor approach that enables the company to capture low-cost CMOS techniques and apply them to RFID devices. Unlike conventional RFID chipmakers, which typically use extra photo masks and process steps to create on-board, non-volatile EEPROM or Flash memories on RFID chips, Impinj engineers use “self-adaptive silicon.” Using the technique, they create special transistors containing gates that are able to store bits of memory. By fabricating such transistors, they can make non-volatile memories without resorting to the extra photo masks and steps required in EEPROMs or Flash.
“Self-adaptive silicon enables us to make non-volatile memory with the simplest of CMOS processes,” says Dimitri Desmons, vice president of marketing for Impinj.
Impinj engineers say that self-adaptive technology can potentially cut pennies from chip costs. Moreover, such technologies reduce costs even further when combined with high-volume assembly methodologies, such as those developed by Alien Technology. Alien, which employs a technique known as Fluidic Self-Assembly (FSA), recently announced it has cut inlay costs to a scant 12.9 cents apiece through its efforts. Its assembly technique, originally developed by a company founder who was playing a child’s game that required him to maneuver steel balls into tiny slots, reportedly allows the company to package as many as 2 million chips per hour into RFID tags, compared to 10,000 per hour using conventional methods. The trick, the company says, is to suspend the tiny semiconductor devices in a liquid and then “flow” them across the holes, where they drop in and self-align. The self-assembly technology, combined with growing production volume, enabled Alien to cut its tag costs by nearly 50 percent recently, down from 23 cents apiece.
Other vendors have attacked the cost issue from a different perspective. Symbol Technologies, for example, has cut the cost of its tags through multiple efforts, not the least of which is its move from silver to aluminum antennae. The company, which makes high-performance RFID systems, was able to move from the higher conductivity of silver to the lower conductivity of aluminum by developing an on-chip “charge pump” that helps boost the continuity and strength of RF signals coming to the antenna.
Symbol has employed the antenna on so-called bag tags used at McCarran International Airport. Integrating two antennae oriented 90 degrees from one another, the RFID tags are virtually assured of being able to communicate with the airport’s RFID readers, no matter how bags are tossed onto conveyor belts.
“If you wanted to have a lower cost solution where one antenna would work, the bags would have to be oriented in a certain way, and they’d have to pass the reader in a certain way,” says Alan McNabb, senior director of product management for Symbol’s RFID tags. “But with our tag, the orientation of the bag doesn’t matter.”
“The pharmaceutical industry has a huge issue with counterfeit product coming through the market,” says Dirk Morgenroth, marketing manager for RFID for Philips Semiconductors. “They’ve been very vocal about using RFID.”
Manufacturers — including Philips, Texas Instruments, Inpinj and Alien — have also landed their RFID products on shirts, pant and sweaters in the fashion industry, as well as in library books, and on DVD and CD cases.
The industry’s biggest score to date, however, could be in the works in Europe, where the European Central Bank is rumored to be working with vendors on weaving RFID into the fabric of its bank notes. The technology — most probably incorporated in larger bills — would enable money to carry its own history. Hence, it would become more difficult for kidnappers to ask for “unmarked bills.” It would also enable law enforcement agencies to “follow the money” in illegal transactions.
The project, originally reported in such publications as Wired and EE Times, was supposed to take effect in Europe’s 2005 currency. Hitachi Ltd., which announced in February that it has developed the world’s smallest RFID chip, measuring just 0.4 × 0.4 mm × 7.5 µm, has often been linked with the Euro reports. Hitachi, however, denies it has worked on such a project. A European Central Bank spokesman also told Design News, “We cannot say anything about this, and we’ve requested that our providers sign a mutual agreement not to talk about it.”
Even if such projects never reach fruition, however, experts are confident that RFID will eventually be the backbone of a plan that researchers have called “an Internet of things,” in which almost everything, large and small, is connected via the Web. The plan, already described in hardware and software protocols, calls for all information on a product to be written in a code based on eXtensible Markup Language (XML). The code, which forms a sort of web page for each item, would be connected via RFID tags to Internet servers. Thus, all products could be identified anywhere, instantly. A broad coalition of corporate giants, including Coca-Cola, International Paper, Johnson & Johnson, Kimberly-Clark, Pepsi, Procter & Gamble and others have supported such efforts through MIT’s Auto-ID Center.
Low cost, of course, is a key to such plans, but researchers have worked that out, too. Ultimately, they say, everyday items will incorporate RFID, not on sticky tags, but through integration into the corrugate of cardboard boxes. Ongoing efforts in this area will be the key to lowering RFID cost, researchers say, because it eliminates the need for certain parts of the tag. Instead of being done as an afterthought, as is the case today, such technologies would be integrated during the cardboard manufacturing process, thus enabling cost reduction.
“These RFID technologies will co-exist with the bar code for a long time into the future,” says Sarma of MIT. “But they will provide information that a bar code can’t. ‘Did the item go to the sales floor? Did the meat sit in the fridge long enough?’ You can’t know that with a bar code.”
Sarma says such technologies will become widespread when production volume reaches a “tipping point.” When that happens, it will drive costs down to a level low enough to motivate use of RFID on everyday items. And with retailers — particularly Wal-Mart — pushing hard for RFID, the concept is not unrealistic, experts say.
“The question now is the tipping point,” Sarma says. “When do you get to the percentage that causes you to say, ‘I’m going to put the tag inside the corrugate?’ In the next year, we could see it happen.”
Reach Senior Technical Editor, Chuck Murray at firstname.lastname@example.org.