Early adopters say the trick to navigating the relatively untested waters of radio frequency identification is to assemble a savvy team that can set the project scope, win over business users and make up for the technology's limitations.
Take, for example, the El Paso County government office in Colorado Springs. A six-week RFID pilot project launched this fall to track IT assets involved tagging 265 PCs, 10 printers and 10 tablet PCs. The first step in the project, which would eventually pour retrieved RFID data into PeopleSoft's Asset Manager, was to create a business case.
Buffy Dorpinghaus, manager of the Colorado county's PeopleSoft group, met with CIO Bill Miller and desktop and network managers. To sell them on the project, Dorpinghaus explained that the new system would use existing bar-code scripts to cut the time needed to track PCs from 10 minutes to one minute without adding hardware or software or requiring a staff member to visit each desktop.
"I had to give them a caveat and said, 'Let's just try it, and if it's not working out, then great, no harm done,' " she says. The pilot would also be cheap, at under US$100,000, Dorpinghaus says. That one-time pilot expense, which also paid for the production license, was a big plus. Had the project been bigger, she says, there probably would have been resistance.
Miller says that although RFID is a relatively new technology for government agencies, the pilot was seen as an opportunity to streamline processes and verify that new software and hardware would work the way El Paso County needed them to.
Next, the nine-person project team was put together. Dorpinghaus was chosen to head the team because of her PeopleSoft applications expertise and ability to map the technology to business processes. Also on board was a non-IT manager, who handled project administration, meeting planning and status reporting, and three IT staffers who would eventually use the system, including one person assigned to transfer the scripts used with the bar codes so they would work with the RFID technology as well.
Jamie Hintlian, a consultant at Accenture Ltd., says this upfront work of building the RFID team with the right skills and creating a solid business case is key. "Make sure the pilot is focused on a common set of objectives and imperatives," he advises. By managing the scope, companies can more easily manage the infrastructure and technology, Hintlian says.
For its RFID pilot, bicycle manufacturer Pacific Cycle created a four-person team: two IT workers with RFID technical skills and two business users who would be using the system once it was up and running, says CIO Ed Matthews. The pilot involved tagging 3,500 bikes and the packaging material inside and outside the boxes they're shipped in.
Since the pilot went live in September, the manufacturer has shipped 15,000 tagged bikes to Wal-Mart Stores Inc. But because RFID technology has been changing so fast, Matthews says that in retrospect, he would have waited another year. "In terms of technology, the immaturity has and continues to be a nuisance," he says. "We tried several tags and antenna designs but didn't get the results we wanted. I can't tell you the number of times that we were told a product is just a couple of weeks or a month away only to find out that it is still coming."
Winning over stakeholders
Another critical part of managing an RFID project is gaining business users' trust. The El Paso team held regular meetings with end users in the help desk and networking departments as well as those in the purchasing and supply group to create a clear, standardized process. The team used Microsoft Corp.'s Visio software to map the workflows.
Creating a structured workflow plan is time well spent, says Hintlian. "The results will be more significant, and the expectations for the pilot will be agreed upon from the start," he says.
Dorpinghaus says the idea was to fully train the workers who would be using the RFID system and ensure that they would benefit from it. Because the end users in this case were IT workers and savvy about the technology, they proved especially demanding, requesting special interfaces to access relevant data, she says.
But the RFID team learned early on that that kind of continued user feedback can only improve the system. "Keep encouraging them to be open," advises Dorpinghaus. Most problems can be easily resolved, and quick fixes can nip any growing dissatisfaction with the system, she says.
"Make sure you have good resource commitments from the pilot participants," adds Hintlian. "Each participant must be able to dedicate enough time to plan and execute well."
In addition, "working together face-to-face is important for the success of these pilots," he says.
Learning along the way
Even though RFID technology has been around for some time, launching a successful pilot requires trial and error, says Sean Clark, director of RFID at DC Logistics, a Dallas-based third-party logistics services provider. His company is launching an RFID pilot project in January for one of its customers, with other tests planned for clients that need to satisfy the requirements of the U.S. Department of Defense and retailers like Wal-Mart.
For example, DC Logistics learned to place a plastic tag under the RFID tag if the container is made of metal so that the signal is read accurately. At Pacific Cycle, several dry runs uncovered the best places on the bicycles to place the tags and whether the tags could be successfully stored and retrieved, says Matthews.
An important skill for RFID team members is the ability to improvise, he says. "If you looked at our warehouse and the antenna setup around the portals, you would think it was a mad scientist gone bad," says Matthews. "To get better read rates and to protect the antennas from being hit from a forklift, we have tried everything except attaching an old TV tray."
In fact, Matthews says bicycle welders were called on by the project team to manufacture mounts to attach RFID equipment to dock doors.
"I do believe that all of this will be a no-brainer in a couple of years," says Matthews. "But for now, they don't call it the bleeding edge for nothing."
That ability to roll with the punches is an important attribute in team leaders as well as team members. "Of course there were mistakes," acknowledges El Paso's Dorpinghaus. For instance, she says she forgot to change the license for the RFID system from development to production when the pilot went live, a problem that was quickly resolved.
She also soon discovered that there aren't a wide variety of readers or tags to choose from. The county had hoped to get more "active" tags -- tags with ranges exceeding 8 feet -- but eventually had to settle on passive tags with shorter ranges. "Be patient, be positive," advises Dorpinghaus.
"Just because it doesn't work the first time, try again," adds Clark. Some tweaking might get the system up and running in the pilot. "You need to tinker at this stage," he says.
MANAGE PROJECT SCOPE. If the pilot is focused on a common set of objectives, the infrastructure and technology will be easier to manage.
CREATE A WORK PLAN. A well-structured plan will deliver better results and help meet user expectations.
GET STAKEHOLDER BUY-IN. Each pilot participant must be able to dedicate adequate time to plan and execute on the project.
GET THE TEAM TOGETHER. RFID project teams should work on-site together, preferably face to face.