Ready or not, here it comes. Radio frequency ID technology in 2004 will become the next big thing in supply-chain management for food manufacturers, largely because Wal-Mart Stores Inc. wants it to be. Whether Wal-Mart, its suppliers or the technology will be ready to perform when the switch gets flipped, however, remains a question.
Doubts about the technology and the supply chain's ability to deliver on it have forced Wal-Mart to scale back its ambitious agenda to have its top 100 suppliers, including just about any food company with $3 billion or more in annual U.S. retail sales, put RFID tags on cases and pallets by Jan. 1, 2005. What was to have been a national rollout has been scaled back to Texas initially.
Opinions on RFID vary widely, from those who see it as an efficiency improvement on par with uniform product codes (UPC) to those who consider it an unprofitable burden for manufacturers of food and other fast-moving, low-price products. Even RFID naysayers, however, say resistance is futile. With Wal-Mart and the U.S. Dept. of Defense already on the RFID bandwagon, few major food companies can afford not to hop on with them.
RFID pros and cons
RFID chips potentially will allow Wal-Mart and its suppliers to track pallets and cases of product through the supply chain to the shelf, helping reduce out-of-stocks, labor costs and inventory. RFID chips have the capacity to track not only individual stock-keeping units (SKUs) but also each individual item using an electronic product code (EPC) that appears destined to replace the UPC.
Ultimately, having RFID chips on individual products -- which isn't broadly expected to be practical until the end of the decade -- could further reduce out-of-stocks, streamline checkouts, and vastly improve retail sales data. They could enable what Procter & Gamble Co., one of the biggest packaged-goods backers of the technology, labels "the consumer-driven supply chain," where actual purchases, rather than forecasts, drive the process.
"Even with the barcode, there are errors in terms of misreads or human error," says Ralph Drayer, principal in the Cincinnati-based consulting firm Supply Chain Insights and former chief logistics officer at P&G. "With RFID, you're going to have a much more accurate picture in real time of what's happening in the supply chain."
Beyond eliminating errors, RFID also can eliminate labor associated with scanning barcodes and can improve product traceability, which will help in case of recalls. The technology may become a key part of food industry compliance with country-of-origin labeling in the U.S. and similar requirements in Europe, according to Forrester Research, Cambridge, Mass. It's also been suggested as a method for tracking cattle in the U.S. in light of the recent mad cow discovery in Washington state.
Reducing out-of-stocks and sales lost because of them is likely to be the key financial benefit for suppliers, Drayer says, noting much of that benefit will have to wait until it's practical to put tags on products. But he says benefits do start at the pallet and case level in the distribution center, noting that P&G already has been using the technology in its warehouses to better track products.
"There's no doubt in my mind there are huge opportunities [with RFID]," says Joe Andraski, former logistics executive with Nabisco and now senior vice president of OMI International, Dallas, a provider of supply chain execution and e-commerce software and services. He also is managing director of the Voluntary Industry Committee on Standards, an industry group that develops standards for data transfer and collaboration practices. "We're still dealing with out-of-stocks at retail on the level that Coca-Cola found 10 years ago. We're still dealing with spoilage, which actually has increased over what it was 10 years ago."
Still, Andraski says, "I think the jury is still out in terms of all the financial paybacks that companies are going to experience. You're going to find people who have not been satisfied with the results of their investments to date in supply-chain technology and are going to balk at investing any more, especially in something that's unproven. As always, there's going to be that 10 percent who are eager and a whole bunch of people who will wait and see and follow."
Following has its rewards
Following, in this case, may not be a bad idea. The key variable cost in RFID is for the tags themselves, now costing around 15 cents but expected to fall to as low as a penny by the end of the decade.
Indeed, Chicago-based consulting firm A.T. Kearney recommends in a recent report food companies drag their feet as long as possible on implementation, though they should begin preparing their systems and people for implementation as soon as possible. The firm projects a 12-18-month delay in implementation could save a $5 billion food company more than $100 million.
But since most of the benefits for food companies will come only if trading partners share inventory and sales information to reduce out-of-stocks, the firm says it's important to start discussions on data sharing immediately.
"Plunging headfirst into a full-bore implementation is not economically viable," according to the report. "Nor is waiting for the industry to fully embrace the technology and systems before starting an implementation effort."
"A lot of manufacturers faced with an ultimatum simply will slap an RFID tag on packaging and never really get the benefits they could be getting if they invest in the process changes [that should accompany RFID]," Drayer says. He compares the situation to continuous replenishment, which became a retailer mandate in the 1990s. Many companies didn't make the internal improvements needed to take advantage of just-in-time manufacturing, and instead increased their costs by building an inventory buffer to meet retailer demands. "RFID," he says, "has the same potential."
Rich , and retailers , get richer
There's a good reason for some suppliers to be more enthusiastic about RFID than others, notes A.T. Kearney. Suppliers of high-value items such as drugs, electronics and cosmetics will get far more benefit from RFID , such as reduction of theft , at far less relative cost than will food processors, the report notes. Not surprisingly, among the biggest backers of RFID in the consumer packaged goods world have been P&G, Gillette Co. and Kodak, makers of higher-ticket, higher-margin products.
Most food companies, notes A.T. Kearney, already have highly efficient supply chains, so they won't see significant net benefits from RFID. Yet their costs will exceed those borne by makers of higher-ticket items.
A grocery products manufacturer with sales of $5 billion will require 221 million tags for all pallets and cases produced annually. That would mean $33 million a year just for chips alone at the current 15-cent price, falling to $11 million if prices fall to 5 cents. By contrast, an over-the-counter drug manufacturer of the same size would see annual costs of only $2.2 million initially and $700,000 later.
The primary investment for retailers will be up front, amounting to $400,000 per distribution center and $100,000 per store, plus systems integration costs of $35-40 million. Yet retailers stand to see far more benefit than even manufacturers of higher-ticket items, according to the report, ranging from less labor cost to fewer sales lost to out-of-stocks and less theft.
Meanwhile, Wal-Mart has told suppliers it won't pay extra to defray RFID costs, says Kara Romanow, analyst with AMR Research, Cambridge, Mass. And under the Robinson-Patman Act, manufacturers will likely be prohibited from charging other retailers for something Wal-Mart gets for free.
In short, RFID appears likely to shift margins from manufacturers to retailers and from food processors to makers of higher-margin products.
There may be something of a Robin Hood effect, however, within the food industry. While bigger food companies with advanced supply chain management systems won't see that much internal cost savings from their RFID investment, smaller companies with less-advanced systems may see bigger savings, according to the Kearney report. Smaller companies also will pay less for RFID chips, since Wal-Mart doesn't expect suppliers smaller than the top 100 to tag cases until 2006, by which time tag prices should fall by as much as 67 percent.
Hype yields to confusion
Not since the Internet has a technology been so breathlessly hyped as RFID, whose skeptics view the initials as standing for "Ripe For Interminable Discussion." Perhaps not surprisingly, some of the same people have been doing the hyping this time around. Forrester, known for touting the Internet in the 1990s, has been among the loudest cheerleaders for RFID. Forrester and other poetic backers of the technology bill it as the linchpin of "The Extended Internet" or "The Internet of Things."
However, the technology apparently still has a ways to go. It's still impossible to read RFID signals through liquids and metal , eliminating practical use for much of the food and beverage industry at this point, according to Romanow. And powerful as it is, Wal-Mart may have trouble overcoming the laws of physics and the limitations of current RFID technology. Wal-Mart said it expects 100 percent accuracy in read rates on RFID chips, but recent tests have shown accuracy rates of only 80-90 percent, she says.
By narrowing the initial RFID rollout to 150 stores in Texas, Wal-Mart may have done more to help itself than its suppliers, who may have trouble sorting products headed to other places. Romanow recommends manufacturers initially should use cross-docking facilities in or near the three Texas distribution centers to break down pallets and apply tags.
The Kearney report says suppliers may have to tag products bound for all retailers at the plant level once Wal-Mart requires full national RFID compliance, because breaking down pallets at the warehouse level to tag only Wal-Mart cases is likely to prove too costly.
Still, while Wal-Mart was short on practical details in its November meeting with top suppliers, which Romanow likens to "a pep rally for RFID," she sees positive signs the giant retailer is becoming more flexible. One is that Wal-Mart dropped a requirement that RFID signals on each case be readable when the cases are still on the pallet, which should at least temporarily overcome the problems of reading signals through metal and liquids.
"There were signs of pragmatism," she says. "But some shifts were too subtle for many attendees to interpret as such."