Announced with great fanfare just three years ago, the great Canadian radio-frequency identification (RFID) experiment was if anything ambitious.

At about the same time Walmart was grabbing all the RFID headlines by requireing its suppliers to become RFID-compliant, a coalition of Canadian food manufacturers, grocery associations and a retailer began tagging select food products using RFID tags.

Running for about a year, the pilot was led by the Canadian RFID Centre (hosted by IBM Canada) and included a group of well-known industry names including Maple Leaf Foods, General Mills, Scott Paper, Unilever and Loblaws.
Centre members had hoped a food product pilot would help Canadian Council of Grocery Distributors members and Canadian food manufacturers and retailers assess the costs and benefits of an RFID implementation by providing details about how well different technologies function with different products and in different environments.
Manufacturers applied RFID tags on cases and pallets of select product lines prior to shipping. The goods were then shipped to Loblaw distribution centres and the back rooms of its Toronto-area stores. Stationary and handheld readers were used at the manufacturing sites and distribution centres to read products which included paper goods, dry goods, frozen food, meat and canned goods. 
As well as tracking the shipment of goods, the data created by the pilot was be of great value to the participants. The data collected was to be a gold mine for better distribution strategies leading to improved business operations. RFID was about to hit the computing mainstream, and the grocery industry of Canada was going to lead the way.
But by May 2007, it was becoming apparent that the results were not what had been hoped for.
According to a report filed after the pilot, “The results … showed that for many product categories and packaging types, further RFID read rate improvements are still required before there is rationale for adoption.”
Reading trouble

The initial average read rate for the nine piloted SKUs used was 71 per cent. While average read rates eventually rose to 89 per cent after modifications and adjustments to equipment including the firmware and middleware, the results were not good enough to justify further investment.

“There just has not been a huge uptake in Canada,” says Nigel Wallis, research director for application solutions at IDC Canada. “The expectations about three years ago were crazy, so people lost interest.”
The problem is that RFID is complicated and expensive.
“It means reconfiguring your warehouse:  It’s all the tags and readers and it’s all the hardware equipment,” Wallis said. “On a massive scale, the upfront costs are higher than anticipated, and so it necessarily isn’t going to have the payback.”
“It’s just not well-suited as part of an extended enterprise information system,” says Simon Ellis, practice director for supply chain strategies at research firm Manufacturing Insights, a division of IDC. Ellis blames high upfront costs and complexity of technology for its slow uptake.
Before he joined IDC, Ellis led the RFID program at Unilever PLC, which manufactures a variety of consumer goods, including Pond’s women’s skin care products, Becel margarine, Hellman’s mayonnaise and Lipton tea.
Without question, RFID is well-suited for closed loop or stand-alone applications, where it competes with bar code and magnetic strip technology, says Ellis. Examples often cited are access control, payment cards and personal identification (such as patient-tracking).
However, for distribution centre efficiencies and reduction of out-of-stock situations, once touted as the greatest opportunity area for both retailers and manufacturers to profit from RFID adoption, it’s overly expensive and doesn’t solve problems.

“I just don’t think it is appropriate for item tracking,” Ellis says. “It is overkill.” According to Ellis, there are better solutions for tracking inventory such as retail software which can be used to by capture data at point-of-sale without the high cost of equipment-intensive RFID systems.

Robert Moroz, president of RFID vendor R. Moroz Ltd. of Markham, Ont., agrees that the hype of three years ago did the RFID industry  no favours as “users quickly realized that a lot of things needed to be put in place” before the benefits could be realized.
But although the highly-publicized supply chain projects did not work out as envisioned, he is confident “a whole second wave of RFID solutions” that combine remote sensing and cellular technology will drive a lot of amazing new applications, especially in health care. Doctors, for example, may collect patient data using sensors, then “broadcast” the data to a health care facility.
Moroz, lists a number of RFID’s strengths: “It’s wireless, it’s low cost and it’s compact,” and should benefit as more and more intelligence is embedded in the RFID chip, and as hardware and software suppliers learn to better understand it.
Moroz remains convinced of both the short-term and long-term potential for RFID, although it may be not until the middle of 2011 before the so-called “second wave” of RFID applications come to market.
Also, leading the way is IBM, which continues to invest heavily in the technology. In August of this year, IBM introduced WebSphere-based RFID middleware that can gather data from a wide variety of networked sensors, analyze it, and feed it into other enterprise applications that can use the data to make decisions.
The sensors, the company says, have proliferated in recent years, especially with the advent of RFID tags which can be attached to products in transit or to corporate assets such as computers and hard drives.
“To capture the data in the context of when it takes place, you need software to correlate all that. The back-end into ERP is what the (IBM Websphere announcement) is all about,” says Martin Wildberger, vice-president of sensor solutions at IBM’s software development lab in Markham, Ont.
Adoption will increase

“RFID is a great example of a sensor technology. As things become more instrumented, adoption will increase,” says Wildberger, who, far from believing RFID is only a niche, sees it as having broad application in all industries and as business begins to understand its benefits.

In late September, IBM concluded an agreement with Korean-based steel manufacturer POSCO to use ultra-wideband radio frequency identification technology to monitor all areas in the facility. Everyone working in or visiting the plant is given a tag, allowing managers to respond track personnel quickly in the event of an accident or calamity such as a fire or a gas leak. In the past, POSCO had no way of finding employees.
 “It will take a while before you go across all the standards issues, but the way you attack the problem is to have it more thought out,” believes Wildberger.
But despite those initiatives, there is general disagreement over whether RFID’s future is as a niche rather than mainstream information systems technology.
“It won’t go away, but it will be a niche, a technical tool for some interesting applications,” says Wallis of IDC Canada.
And now that Walmart’s ambitious project to affix radio frequency identification tags to merchandise — which has been the poster boy for RFID for so long — is rumoured to be in jeopardy, skepticism is again on the rise.
Indeed, IDC’s Ellis believes that though not yet announced, it’s almost inevitable Walmart will wind its RFID project down, That setback, not to mention the Canadian grocery industry experiment, indicate it may be a while before RFID technology becomes mainstream and a concern to IT departments.
While clearly RFID has a role to play in specialized applications, broad enterprise-based supply chain solutions are more difficult to find.
Earlier this year, EPC Global Inc., a subsidiary of the global not-for-profit standards organization GS1, completed the third phase of its Transportation and Logistics RFID pilot program.
Early successes in that particular RFID pilot project is proof that technological limitations of RFID can be overcome, and IT departments will sooner or later need to become involved, says Shirley Arsenault, an RFID expert on loan to GS1 Canada from Toronto-based software supplier Allumis. 
Arsenault talks about an EPCglobal RFID trial.

And while there are those who believe RFID seems destined to remain a niche technology for the next few years, its potential application for enterprise wide systems  remains full of promise — even though it may take some time.