In the next few years and beyond, the supply chain is aiming at a major facelift, some of which is already under way. If success favors those who are prepared, then it's a good time to take a look at the current state of radio frequency identification (RFID) as the tool being bet on. For businesses, RFID is simply about using radio waves to automatically identify physical items in varying proximity to machine "readers" which can uniquely identify them at the ship, truck, container, pallet or unique item level.
As it applies to the supply chain, RFID and electronic product codes (EPCs) are not just a replacement for the under-appreciated barcodes which revolutionized the manufacturing, retail and shipping businesses 29 years ago. They are a giant step forward in supply-chain visibility which one day should track goods from raw material to landfill, and simultaneously address issues like counterfeiting, theft, and perishability.
Before you call in the hyperbole police to arrest us for overstatement, understand that it's the availability and sheer variety of applications that get people thinking in Star Trek terms. Consider the supply chain parallels in this real example of technology in use today:
In large marathon races like those run in New York or Boston, officials need to account for not only the winner, but for each of thousands of runners jostling in the streets. At the start of the race, contestants affix a small transponder to their bodies. Machine readers mark the correct time they cross the starting line, chart relative progress at many intervals, reveal where they slowed, sped, or dropped out. Spectators watch the ebb and flow on a leader board, and a permanent record secures the outcome.
That's basically it. RFID may revolutionize the supply chain, but it's hardly a new concept. "Friend or foe" beacons identified military aircraft as far back as the Second World War. Newer uses arose about 17 years ago in livestock and vehicle identification tracking. Lately, salmon roe are tracked in rivers with RFID, and there is talk of RFID tags in every euro bill. Though ubiquitous RFID is no a slam-dunk business certainty, the technology works today in applications where computers are able to recognize things around themselves automatically.
Efficiency, Security, Authenticity
Kevin Ashton is executive director of the Auto-ID Center, a research arm of Massachusetts Institute of Technology (MIT). Charged with building consensus and standards for uses of RFID, Ashton's group might consider how the postal service could improve delivery, or how the Department of Defense might fight a war more efficiently. The business side of RFID, he says, is the real prize. "The question is, 'How do we use this to sell more stuff and be more profitable?'" In Ashton's mind, a warehouse is just an excuse for not knowing what you need.
The center is also working with a bandwagon of businesses (initially in retail and pharmaceutical), that are testing and applying RFID for top and bottom-line benefits. "We got involved in RFID and EPC trying to solve two problems and take advantage of one opportunity," says Larry Kellum, director of B2B supply chain innovation at Procter & Gamble.
The problems are theft and counterfeiting, supply chain failures that are respective $50 and $500 billion problems for global retailers. The opportunity, he says, is the potential to identify products right down to the unique unit level and follow them through the supply chain.
Theft is addressed today with at least four technologies, acoustic-magnetic, (the little white rectangles on high-value goods); foil radio frequency tags; electromagnetic; and microwave tags. While these systems were initially bought and applied by retailers, the cost became such that manufacturers soon took over some of the work, applying security devices inside costlier packaged goods, perhaps working with four piles of inventory.
As for counterfeiting, which many consider an Asian problem, Kellum says P&G has found counterfeit bottles of Head&Shoulders shampoo near its Cincinnati headquarters. "There are three occasions this year where the FDA has pulled pharmaceutical drugs off shelves when there was no way to tell the real from the fake," Kellum says. In such cases, whole stocks were destroyed. In the wake of terror attacks, the scrutiny will only increase.
Meanwhile, product identification remains the work of the ubiquitous barcodes, UPC in the United States, and EAN in 96 other countries around the world. "Barcodes have had almost immeasurable impact," says Ashton, "but they're probably the only piece of information technology from the 1960s that we still use unchanged."
And, barcodes are not really automatic identification technology. Applied by manufacturers, they're usually read once, manually, at the checkout counter. "That's not the only time you want to know where a product is, or what it is," Ashton says. Point-of-sale information is of little use if an inventory system is waiting to sell 10 items that have been stolen off the shelf, and not many businesses can afford more than annual inventories. RFID won't stop street-corner fencing of goods, but it could prevent a retailer from buying gray-market perfume.
Tag, You're It!
RFID/EPC technology takes the security idea a step farther. So-called "smart tags" are un-powered microcomputer chips activated when placed in the transmitting field of a fixed or moving reader. "In its lowest cost implementation, it has just enough information to say its name or shut up," says Ashton. Though it transmits nothing more than a unique number, when connected to a network like the Internet, its value multiplies.
There's plenty of theorizing about the value of "smart shelves" that can itemize inventory in real time, self-checkout, managing expiration dates and item location in stores. "In 35 percent of cases, people walk out of apparel stores without product when the product was there, but the customer or sales rep couldn't find it," notes Pete Abell, director of retail research at AMR. RFID, he says, has many niche applications in higher-priced perishables and shelf goods with expiration dates as well. "It's a value equation," Abell says. "No one wants fines or consumer lawsuits."
The value equation is the central issue around RFID deployment, a mix of product ubiquity, value, and the cost of the tag. A 40-cent tag makes sense for a leather jacket, but not for a can of soup. So, many retailers and middlemen are experimenting with RFID at the pallet, and perhaps the case level. CHEP, a large, London-based company that pools pallets and containers, launched a pilot last year to install 250,000 EPC-compliant chips on its products. A company like Wal-Mart, big as it is, has only three pallet suppliers. Pushing the scale to 500,000 pallets brings the cost down; Abell believes Wal-Mart is hoping the technology, even at that level, might lower supply chain costs 3 to 5 percent.
P&G is already in a pallet-level field test with a Sam's store in Tulsa; next February it begins case-level testing with Sams and Wal-Mart on products shipped from P&G's Missouri plant. If they choose, testing could include active, self-powered tags connected to GPS units, and even temperature sensing tags. Reconciling damage in transit and invoice disputes are still more potential benefits of RFID.
In early 2003, P&G also plans tests at the item level, and will even include some smart shelves that track the identity of unique bottles or cans, not just grades and classes. But making RFID practical on the item level is still a ways out, a chicken and egg problem that will determine if quantities ever drive chip cost to a nickel or less. "If we tag all the pallets and cases we should in a year, that would take about 2.5 billion tags," says Kellum. "At the item level, it's more like 22 billion. Until you build five billion chips, you can't get to a nickel."
All Tags Equal
The P&G case is just one of scores or hundreds of simple and deep trials underway in the U.S., and to an even greater extent in Europe. Many of these projects are under wraps at companies that feel they're protecting a competitive advantage. The industry and solutions are fragmented to the point where Kellum has identified 123 protocols for RFID. "Just about all of them work, and they're all incompatible," he says.
On October 1, 1999, (the 25th anniversary of the UCC code), the Auto-ID Center came into official existence in MIT's engineering department. Having developed the barcode in the first place, MIT was a logical setting to extend the discussion. Funded by UCC, Gillette and P&G, the center today has 50 sponsors on three continents, a who's who of CPG, technology, shipping, and retail interests, not to mention the U.S. Department of Defense and Postal Service. A work in progress, the Auto-ID Center sponsors have met, developed and demonstrated technology, and worked for a common set of standards for RFID.
But there were bound to be squabbles, such as those over extant UPC and EAN standards. To their credit, the group has opened Auto-ID centers in Cambridge, England, and Adelaide, Australia; next month, new centers will be announced in Tokyo and Shanghai; another is planned in South America.
The actual code structure may not be the greatest issue. "Revenue is the dirty little secret," Abell says. Both EAN and UPC are paid for issuing code to suppliers in their countries. While UCC governs 26 industries in North America, EAN is fragmented across 96 nations. "They all have little bureaucracies to feed," Abell says. As a result, the Auto-ID Center has made sure it can set aside enough unique digits so all the parties can continue to issue code.
Also, technology is ahead of the game. "Relative to standards, we always say, 'Look at whether RFID can improve your automation today and go from there,'" says Susy d'Hont, marketing manager at RFID system manufacturer Matrics, one of some 30 tech partners welcomed but treated neutrally at the Auto-ID Center. With technology a proprietary advantage, best-practice leaders won't wait for consensus. The process is incremental, though d'Hont says without standards, the market might never scale to the billions.
There are technical issues as well. Different countries allow different frequencies and power levels for RFID devices. For Matrics, 10 feet is the minimum standard for RFID measurement; other system vendors see it differently. The very properties of the materials being scanned, plastics, liquids, metals, can also affect the properties of the devices.
Finally, you might also worry that ubiquitous use of RFID would make the tags themselves subject to counterfeit. With proper infrastructure in place that would be hard to do, Ashton says, the equivalent of typing a Web address into a browser for a page that doesn't exist.
A Meeting of Minds
The proof will be in the pudding. Mark your calendars for October, 2003, the date set for a symposium in Chicago. There, Version 1 of EPC automatic identification will be rolled out, with open specifications for tags talking to readers, readers talking to computers, standards for capturing and managing the data and so on. Hopefully, a lot of vendors will be present as well. "People will be fighting to sell you compliant tags just like they fight to sell you PCs today," says Ashton.
It's not an all-at-once proposition, but Ashton hopes some of the rest of the world will be ready to at least dip their toes in the technology. The more people who use the technology, the more valuable it becomes, and the cheaper it gets. "We'll have six months to a year," Ashton says. If confidence and momentum remain high, things could happen quickly from the end of 2004. "By 2010 it could be a whole different world or nothing could happen," he says, with a scientist's aplomb.
He's also hoping for results beyond pure capitalistic efficiency. The Auto-ID center is working with consumer and watchdog groups and promises easy, secure opt-out for consumers who fear obtrusive marketing and data gathering. In the end, he feels you're identifying a product, not a person, and for those who participate, the technology will be no more intrusive than a grocery loyalty card. For futurists who see the days of "smart appliances," refrigerators or microwave ovens that can interrogate their contents, the technology may be helpful, but that's way down the road.
We have moved past the starting line though. Recycling, Ashton says, could be the greatest application of all. Imagine if, in 20 years, everything had a chip in it. The landfill operator could establish what he has and what to do with it. "So many things go into holes in the ground because we can't sort them out."
Will we ever reach the day when every product made carries a unique number? Are there even enough numbers to go around? Leave it to MIT to at least get that issue out of the way. EPC Version 1 contains 96 bits (ones and zeroes). "Fifty-six bits is enough to number every grain of rice consumed on the planet this year," Ashton says. "One hundred twenty-eight bits can number every molecule on the surface of the Earth." In this regard at least, future-speak or not, Y3K is already taken care of.
(Jim Ericson is editorial director and senior news editor at Line56 Media)
source: http://www.line56.com/articles/default.asp?NewsID=4025 5jun03
|
If
you have come to this page from an outside location click
here to get back to mindfully.org |