Retailing is on the verge of a revolution in terms of how it tracks products from manufacture to final sale, thanks to radio frequency identification technology (RFID), which uses tiny microchips in tags to replace bar codes. But some retailers are concerned that the revolution is being stalled because worldwide standards for the technology are still some way off.
RFID chips are radio devices that emit a unique ID when scanned at close range. When pallets, crates and individual products are tagged, scanners linked to logistics databases can track goods with great accuracy. The result should be less wastage and loss in the supply chain, and more accurate stock checking and replenishment at stores.
As being tested by the likes of Wal-Mart and Proctor and Gamble in the US, and Tesco and Marks & Spencer in the UK, RFID is mainly being used to make the supply chain more efficient: projects that involve the large-scale tagging of individual product are more likely around 2007. But the tests, though broadly positive, have revealed problems over accuracy rates, reader range and difficulties in linking to existing product databases. Analysts from the Meta Group research company report that half of the test projects have been abandoned.
Real progress will only come when the retail industry is able to pin down standards for RFID technology in the way it has for bar codes - which are now taken for granted worldwide.
At a recent RFID conference in London, Tesco announced it would delay asking suppliers to put RFID chips on pallets until next year, blaming the slow development of standards.
Marks & Spencer, which has tested tagging in items of clothing, says it shares this concern. "There is a degree of separate development between Europe and the US and we have to be careful there isn't incompatibility," said James Stafford, head of RFID at M&S. "Standards are essential for driving the cost down, improving quality through competition and increasing choice of suppliers."
The problem is that RFID is far more ambitious than the bar code. Standards are needed for use of the radio frequency (known as the "air interface") in each part of the world, and for the coding system applied to tags.
The most likely outcome for the air interface is the acceptance of the UHF Generation 2 standard proposed by 13 chip vendors including Philips Semiconductors and Texas Instruments in April. Three other protocol proposals are also under consideration by the industry standards group, EPCglobal, a joint venture of the standards bodies EAN International and the Uniform Code Council. It intends to ratify a standard by September.
But with standards, the devil is in the detail. For example, the proposed standard does not resolve the issue of whether readers should be smart devices running software (favoured by companies such as Microsoft) or dumb appliances that just capture data and deliver it to the network.
In Europe, the use of UHF wireless is restricted, and its use for RFID applications is under the jurisdiction of the European Telecommunications Standards Institute (ETSI). According to John Falck, chairman of TG34, the ETSI group responsible for RFID standards, proposals relating to UHF Generation 2 (known simply as EN 302 208) are moving through the committee stages. He reckons they have a good chance of being fully approved by the end of the year, although the power (wattage) and functionality of the readers will vary between North America and Europe.
He admits that standards in Region 3 - which includes China and Japan - may take longer, but remains optimistic. "There is likely to be critical mass with Generation 2 and that will exert a very great pressure on Region 3 to comply."
Not surprisingly, there is concern that unless ETSI standards are compatible and concurrent with America's Federal Communications Commission (FCC) standards, retailers would incur increased costs by having to use tags of two frequencies.
To add to the complexity, there are intellectual property considerations. Andrew Osborne, director of policy and research at E.centre, the UK marketing arm of EPCglobal, says that as a result, the standards process takes more time than some people want. "You have to be sure that no person or group will make a claim on the standard once it is developed, and it takes time to do patent searches."
Some observers remain concerned. "There is a real fear that there will be different standards in the US and Europe: it could be a VHS vs Betamax situation," says Alan Lawson, research analyst at Butler Group. And even EPCglobal has expressed concern that some retailers want the emerging standard to use "active" tags, which are more expensive than the more common "passive" tags currently in use. The advantage of active tags is that data can be amended or deleted during the product life-cycle.
Meanwhile, according to Meta Group, the fear of choosing what could be the "wrong" standard is preventing companies from launching pilot projects. Meta Group also says the standards debate is being used as an excuse for delaying development by enterprise software vendors.
But there may be unexpected advantages. Paul Stam de Jonge, the director responsible for RFID solutions at LogicaCMG, says that with standards in place, Europe is likely to overtake the US in RFID adoption. "The US approach is largely mandatory - Wal-Mart, for example, is mandating its suppliers conform by a set date - whereas in Europe, a more collaborative approach is reaping benefits," he says. De Jonge believes the biggest hurdle is the reliability of the tags. "There is currently a 20% failure rate. That has to be improved before tags become a commodity."
De Jonge also doubts the emerging standard will have the backwards compatibility leading vendors want. "It is difficult and expensive to develop readers that have a multifrequency feature."
Despite criticism, the momentum behind RFID is likely to make progress faster than that of the bar code. "I've never seen a standard delivered on time: there is always a significant difference between the specification and adoption," says Tom Friedman, head of the US-based Retail System Alert Group. "Clearly, there are political considerations of implementing this technology, and it is important to gain user confidence, which takes time. The bar code took 25 years to get where it is today."
Others point to the fact that RFID depends on a huge data infrastructure to store and respond to the massive amounts of data captured when the supply chain and in-store products are tracked. This requires the deployment of software and databases that are now only beginning to emerge. "It's going to be a field day for the storage vendors, you can guarantee that," says Friedman.
