Recharging RFID: An Interview with Bill Hardgrave
By Frank Quinn, Editor -- Supply Chain Management Review, 3/1/2007
Three years ago when Wal-Mart announced its famous supplier “mandate,” the business press was buzzing with stories of RFID (radio frequency identification). But after that initial surge of attention, which lasted about a year, the buzz seemed to die down. Companies began to step back and say, in effect, “Hey, what's in this thing for me?”Bill Hardgrave is working to get the RFID momentum back... and he's succeeding on several fronts. As director of the RFID Research Center at the Sam Walton College of Business at University of Arkansas, Hardgrave is leading multiple research efforts into the use of this emerging technology. As part of its charter, the Center operates a 10,000 square foot lab that investigates RFID applications in a real-world setting. In September 2005, that facility became only one of four labs worldwide to receive Performance Test Center accreditation from EPCglobal Inc.
Hardgrave believes in sharing the RFID Research Center's work with the greater supply chain community. The research is freely available on the Center's web site. And in the past year, more than 1,000 visitors from 500 companies have visited the Center in Fayetteville, Ark. (For more information, visit http://itri.uark.edu/rfid)
For Hardgrave, RFID is a business inevitability. The benefits are so compelling, he believes, that RFID soon will be seen as a basic necessity of doing business—the “ante to play the game.” Enlightened by the work being conducted at the RFID Research Center and other institutions, more and more companies are reaching the same conclusion. As this happens, Hardgrave says, the momentum for RFID can only build—once again.
SCMR Editor Francis J. Quinn recently conducted this interview.
Q: Tell us about the background and the mission of the RFID Research Center.
A: It really started a little over three years ago when I had the opportunity to work with Wal-Mart's RFID team. This was not long after they had made the very public announcement about wanting their top 100 suppliers to begin tagging products by January of 2005. Wal-Mart made that announcement in June of 2003, and I started working with them in early 2004. It was a great opportunity as an academic to get in on something that Information Week and others have termed a “disruptive technology.” We saw an opportunity here at the University of Arkansas to really be a leader in terms of taking the technology and applying it. So, a single research project grew to what we believe today to be the foremost research center in investigating ways to use RFID to create business value.
Q: There's a lab associated with the Center, right?
A: Yes. The lab is located within the distribution center of one of our sponsor companies, Hanna's Candle Co. This lab is a live working environment—a “dirty” lab. We don't try to control anything. We pride ourselves on the ability to do applied research on how RFID really works. We have about 15 faculty members from across the campus—from Animal Science, Political Science, Computer Science, Engineering, Logistics—and about 100 students involved in RFID-related projects and classes.
Q: What was the thinking behind bringing together these different disciplines?
A: When you're investigating RFID, you can run into many challenges…like “why aren't we able to read these boxes?” So, we bring in Engineering faculty to help us understand that portion of it and maybe learn how to deploy the technology better. When we're investigating how to use the technology in food products, that brings in the Animal Science or the Food Science people. With questions about software, we engage our Computer Science colleagues. Basically, we've put together a collection of relevant capabilities, just as any industry person would do in putting together a multidisciplinary team to attack a problem.
Q: Could you talk about some of the specific research programs that are under way at the Center?
A: To give a little history, after Wal-Mart, Department of Defense (DoD), Metro, Tesco, and others started getting their suppliers involved, there was pushback saying, “Where's the business value in RFID? Does this stuff really help, or is it just costing us?” So we looked across the board and asked, ultimately, what are we trying to accomplish with this technology? The answer: Get the product to the shelf when the customers want it, where they want it, in the condition they want it. One of the best ways to measure the when/where portion of that question is by examining out-of-stocks. We all know out-of-stocks are a problem. We know that, on average, when the consumer walks into a retail outlet, one out of 12 items on their list is not going to be on the shelf. That's really not acceptable. We felt that RFID could substantially reduce that number. So working with Wal-Mart, we did the largest out-of-stock study ever conducted. It ran for six months and included 24 stores, 12 of which had RFID, 12 of which did not.
We scanned every item that was out-of-stock in those stores, not just things that were RFID tagged at that point. We found that RFID reduced out-of-stocks by 16 percent overall—a huge difference. Refining the analysis by sales velocity, we found that for items that sold between 0.1 items per day (or once every ten days) and 15 items per day, out-of-stocks were reduced by about 30 percent. That's phenomenal! It translates to a potential real sales lift of 1 percent or more—and a 1 percent sales lift is huge!
The out-of-stock study was good for the industry because it was the first real tangible piece of objective evidence that said, look, RFID can make a difference. It was done from an academic perspective; that is, we controlled the experiments, and we did it right.
Q: How about other research efforts?
A. We are currently focusing on three main areas: supply chain optimization, cold chain, and item-level tagging. The out-of-stocks study just described was part of our supply chain optimization focus. In this area, we are also looking at what we call “total asset visibility” and “nested visibility.” Total asset visibility refers to the end-to-end supply chain. Nested visibility means seeing an item inside the case, on the pallet, in the container, on the transportation vessel, and so on. In both cases, we're looking into the opportunities that this visibility affords.
We're also doing a lot of research into RFID in the “cold chain,” that is any environment that has to be controlled—cold or hot, for that matter. With frozen foods or fresh flowers, for example, the slightest change in temperature can affect the quality of that product. So we're now using environmental sensors to detect and record that temperature all the way along. We can get great temperature profiles, which gives us insight into what's happening in the supply chain. That information allows us to take a pre-emptive approach in keeping problems from happening in the supply chain.
Finally, we're looking into RFID at the item level. I don't think we're going to see RFID tags on individual cans of soup any time soon. But I do think we will see item-level tagging on products like consumer electronics, apparel/footwear, and big-ticket items. It makes sense for some of those industries to actually leap-frog fast-moving consumer goods industries that are already using the technology on pallets and cases.
Q: How is the research being put to use in the real world?
A: We have helped many companies from a technology perspective understand what tag works best and what tag placement is best for their product. They'll send products to our lab and we'll help them address these questions. We've had numerous companies that have taken the work that we've done in the lab and applied it directly to their day-to-day operations. Our research on out-of-stocks and other areas has been used not only by Wal-Mart and other retailers but also by suppliers. It would be hard to specifically identify companies that have taken our research and put it in play because once we release it to the public, we hope that it benefits as many people as possible. By the way, we release the results of our research to the public via white papers available for free download from our Web site.
Q: Let's talk about benefits. What would you characterize as the key benefits of RFID realized to date?
A: You could characterize all of the benefits that we've seen so far under the umbrella of visibility. RFID has allowed us to see things in the supply chain that we've never been able to see before. It opens up that black box, if you will. Take the example I mentioned earlier; how were we able to reduce out-of-stocks? Well, by tagging a case, we could see when it came into the back room and whether or not it went to the sales floor. That newfound visibility alone improves the pick list process tremendously. Providing that level of visibility in an efficient manner has not been possible before.
We've also seen real benefits in the area of promotions. Promotional items are married to a specific time period or a specific event. For example, DVDs are released every Tuesday. If they're not on the sales floor Tuesday morning, you miss out. Or let's say you're running a Super Bowl ad. If that product is not out there in the middle of the aisle on the day of the game, you miss a great opportunity. The visibility afforded by RFID tells you whether or not the promotional items are on the floor and ready for sale. Research tells us that all too often the items are sitting in the back room, which obviously means that you've lost a sale.
The flip side of this is getting the product on the floor too early. By the time the promotion ads run, you're sold out. In any case, whether you're early or late in getting product on the floor, you have unhappy customers.
Q: How about some of the potential benefits of RFID, the ones we haven't yet realized?
A: I'm convinced that the greatest potential benefit ultimately lies at the item level. That's probably where we lose sight of the product the most, and therefore it offers the greatest opportunity for improvement. In fact, if you can see what you have in inventory at the item level in real time, the potential is phenomenal. At present, all of our inventory models are based upon a perpetual inventory—a best-guess, if you will. But research has shown that as much as 65 percent of those perpetual inventory counts are wrong. That means you're basing your ordering and replenishment decisions on data that's wrong 65 percent of the time! If we could improve that to 95 percent correct, for example, we would see great improvements in the accuracy of our lead time calculations, forecasting models, safety-stock holdings, etc. Think of the tighter collaborative planning and forecasting we would have with our retailers and suppliers. That's all possible once we get to item-level tagging.
Q: Are there any pilots now under way on item-level tagging?
A: Right now, we're expanding our lab by 2,000 square feet to establish a model store. The setup will mimic a traditional supermarket and an apparel/footwear retailer. We are working on a “contactless” checkout system where all items will be scanned automatically and tallied, ready for payment. If you can just walk out of the store without standing in the checkout line, that's Utopia for a consumer! That research is moving along very rapidly. By the end of the year, you'll really see some good data coming out of item-level pilots from us as well as others.
Q: Do you sense that people are finally starting to realize what RFID can deliver? Put another way, is interest in the technology gaining momentum following the let-down after the initial Wal-Mart and DoD surge?
A: Your characterization is right on. We did have an initial surge from the mandates of Wal-Mart, DoD, and some others. Then it kind of leveled off. I think that a couple of things were happening over that period. One is that after the initial activity, a lot of the suppliers to the retail sector stepped back and said, “OK, what's in this thing for me? What are the benefits for our company?”
We also saw some consolidation in the industry that affected the adoption cycle from a pure business case. But I do think that the slowdown is over, and right now there seems to be new momentum. Wal-Mart is using RFID at 500 stores, and it's going to be 1,000 stores very, very soon. They already have announced plans to roll out RFID for all stores, all suppliers, all SKUs within the next few years. That's significant because up until now many of the other retailers were playing the fast follower or the laggard. Now, no one can afford to wait and see anymore, because if you wait much longer, Wal-Mart and the other early adopters will be so far out in front of you it will take years to catch up. So I expect to see a lot of companies move very rapidly on the RFID front this year.
Q: Many companies report running into obstacles when attempting to implement an RFID program. What are the biggest obstacles that you've observed and how do you overcome them?
A: One of the biggest obstacles revolves around deployment of the technology. For many companies, it's still too hard to deploy. I'll use our own lab to illustrate the point. We have very smart, very experienced people working here. But when we get a new reader to test, it may take us several hours to set it up correctly and start capturing data. So, how does a small- or medium-sized business that doesn't have the technology specialists that we do, get their system up and running in a timely fashion? The answer is, it is difficult to do by themselves. We need to get to the point where it's true plug-n-play at Microsoft installation speed. The small business owner needs to be able to take the reader, plug it into his network, have it automatically configured, and then put it to work. The technology has to become easier to use. The good news is: We are getting there. The hardware and software companies are responding favorably to this dilemma.
Another big hurdle relates to the data. RFID is providing data to companies that they've never had access to before, and they're really not sure how to use the data. Imagine getting potentially millions of records thrown at you on a daily basis. It's like drinking from the fire hydrant, too much of a good thing maybe. That's why there's a growing software industry to help people deal with RFID data. And as it continues to grow, you will start to see these data obstacles removed. But right now, a basic lack of understanding of how to use the data remains one of top barriers to adoption.
Q: Going back to that plug-n-play notion, is that capability on the near-term horizon?
A: Microsoft already is working on a solution. In fact, I saw an early version of their plug-n-play about a year ago. So I'm hoping later this year we'll see something from them that makes RFID truly plug-n-play.
Q: Making a persuasive business case for RFID is a crucial task for our readers. Walk us through your recommendations for doing that.
A: I use a simple model (see Exhibit 1 on page 44), but I think it works well. The first principle is that with RFID—and really, any new technology—it's more than just about putting the technology in place. You put the technology in place because it's going to produce some data for you, which is going to provide a level of visibility you've never had before. So you deploy the technology, generate the data, and then use the data to create business value. That's the sequence. But it's not a linear path because all along the path you have feedback loops. That is, as you deploy the technology, you produce new data, which might give you some insight into how you can better deploy that technology. So it loops back there. Then as you gain insight from that data and it helps you create business value, you might identify additional data needs. So there are more loops.
Looking at the model from another perspective, you start with the pain points—that is, potential business cases—and figure out what you could be doing better. That's your big opportunity, your low-hanging fruit. That's the area where RFID might be able to help. Then you say, all right, what data do we need to help us achieve this opportunity, and then what RFID technology do we deploy to help us with that? Ultimately, it is about one thing: creating business value.
Q: Who within an organization should lead in the development of the business case for RFID?
A: It has to be a multidisciplinary team. RFID is not an IT project, or a supply chain project, or a manufacturing, or a store operations project—it's all of the above.
Q: It's a business project.
A: Right. And how do you attack a business project? By pulling together those key areas from the business. The successful pilots so far have been guided by interdisciplinary teams. A company can't look at RFID and say, “Oh this is an IT issue, so we'll give it to the IT department to handle. Or this is a supply chain problem, so we'll let them figure it out.” I heard a presentation by the CIO of Best Buy who succinctly characterized it when he said that RFID is a customer issue, it's all about serving the customer. When you attack it from that perspective, then the retailer and the supplier both win.
Q: Is top management at the retailers or suppliers starting to recognize the potential in RFID?
A: I'm seeing movement toward that, but we are still mostly seeing the mid-level managers, the director of supply chain or director of IT, coming in to visit us at the lab. Now it could be that the interest started at the top and got pushed down. But I would like to see more C-level people, more VPs, coming in and visiting with us.
Q: Let's discuss some of the applications for RFID in supply chain security. What do you see as RFID's role in this space?
A: There is a natural fit with border security in putting RFID tags on containers coming in on ships. Right now, less than 5 percent of the containers coming into U.S. ports get physically inspected. If RFID technology could enable us to see in real time that a container had not been tampered with from its origin, then that would have significant safety and security implications. From a border-security standpoint, it would be truly phenomenal.
Q: What about other security implications?
A: RFID will really improve security in other important areas as well. For example, the Food and Drug Administration has already said that you've got to be able to show ownership of drugs all along the supply chain, from manufacturer to the consumer. RFID is the perfect technology to support that. Counterfeit drugs are a multi-billion dollar industry. If you can reduce that by even a percentage or two, that's a huge improvement. RFID technology will be extremely helpful in that arena.
We're also going to see more and more use of RFID in food tracking, which can address the kinds of problems we've seen lately with e.coli and other contaminated foods. If we can trace the history of a food product from the farm to a particular market, that would produce tremendous benefit. Without RFID, for example, if you found that you had e.coli from this one farm with this one type of lettuce, you would have to trace all shipments of this lettuce from all of the affected distribution centers (DCs) and the hundreds of stores they may have gone to. You don't have any choice but to alert all those consumers in all those areas and possibly create a panic.
Now assume that we had RFID. We could track the individual cases to the DCs and to the specific stores. You can identify and isolate the contaminated product and determine that it only went to three stores, for example. So instead of having widespread concern and maybe even panic, you can focus your efforts on these stores. RFID can really make a difference in such cases. Coupled with the use of sensors and monitoring devices for temperature and humidity, we can minimize—though probably not totally eliminate—those types of problems before they happen.
Q: What do you see as the next big technology breakthrough in RFID?
A: I see a couple of things ahead. Until now we've had mostly very static readers—those big portals that were set around dock doors. They worked well, and there's still a need for them. But what we're going to see going forward—and in fact have already started to see—are mobile readers that will continue to get smaller, cheaper, and faster. For example, in the very near future, I envision an RFID-enabled wearable vest. A store associate would simply put on the vest and walk up and down the back room to take inventory.
Mobility is going to be key, and it will extend to so many other things that we haven't even thought about yet in terms of analyzing product movement and patterns. Of course, we've done time-and-motion studies for years, but those are snapshots. Now think about being able to do time-and-motion studies where the time frame is limitless.
Before long, we'll also see chipless tags where the tags are simply printed using electromagnetic ink without silicon chips. So instead of printing a bar code on a can of soup, we'll be printing an RFID tag on it. That gets us to item-level tagging.
I think we'll see a convergence of all these technologies coupled with things like global positioning devices into a total wireless sensor network. All of these technologies will really start working in conjunction with each other, and you'll have visibility and insight that you've never had before. Until now, these technologies have been working independently in silos.
Q: What about the privacy issue surrounding RFID? Is that still a potential barrier to growth?
A: The privacy issue is a very touchy one. We have a tendency sometimes to overreact to new technologies, and we want to punish the technology rather than punish the people who are committing the bad acts. Ultimately, it's about the information. We have many laws and regulations protecting information, as well we should. We need to focus on people who violate those laws, rather than the technology that's producing the information. Certainly, we need to be aware of the concerns raised by the privacy advocacy groups because these need to be discussed and we need to make sure the potential problems they're worried about don't happen. A lot of what they think will happen is science fiction, but you never now how quickly technology will develop.
In the long term, I don't believe that privacy will be a major issue. If you roll the clock back 30 years and look at the introduction of the bar code, some groups were worried about invasion of privacy. Some people would not buy items at the supermarket because they had bar codes on them. It's the same thing all over again today. What really matters to most people is getting the products they want on the shelf when, where, and how they want them. They're not really going to care whether the system that does that uses RFID or bar codes. After people realized the benefit of bar codes, the privacy concerns went away. I think we will see the same thing with RFID in due time.
Q: Put yourself in the place of a typical Supply Chain Management Review reader. You might have a logistics or purchasing background and instinctively believe that RFID is something potentially very important to your company. But you feel a little inadequate in terms of understanding the technology and what it can do. What can you do to make yourself smarter about this technology?
A: Let me relate an anecdote in answering that question. I'm an information systems professor, not a physicist. When I first became involved in the technology, I met with the CIO of a Fortune 500 company who had been working on RFID for some time. During the course of our lunch I told him that I was a bit concerned because I didn't have a background in RF technologies. He said, “Give it three months and you'll be one of the foremost experts in the world.” His point was that it's not all that hard to learn about RFID. The knowledge base out there is fairly shallow. It doesn't take a lot to get caught up on the great work that many have already done.
The advice I would give to people trying to figure this out for themselves is to take advantage of the information that's out there. There are some great publications like RFID Journal. There are excellent conferences throughout the year dedicated to the subject.
In addition, people seem to be very willing to share information on this technology. So call on your peers from other organizations that are working on RFID. It's an interesting technology in that not one person or entity can hoard all the knowledge and hope to gain competitive advantage. It requires a critical mass, just like the Internet or cell phones. If not enough people are using the technology, then nobody is going to gain any advantage from it.
Q: Any other good resources?
A: Look to established organizations such as EPCglobal and AIM Global for information. Also, more and more universities are looking into the technology. At the University of Arkansas, for example, one of our main objectives is to help educate the public. We offer courses and training in RFID. Also, in the past year, we have had more than 1,000 visitors from 500-plus companies come through the lab. We welcome anybody to come into our lab, take a look at the technology, and sit down and talk with us about what RFID might mean for them and their company. So there are a lot of avenues to educate yourself about RFID. And now is the perfect time to take advantage of them.
Q: Any final thoughts for our readers?
A: Before too long RFID will simply be the price of admission to play in the business arena. Simply having RFID is not going to provide a competitive advantage, it's going to be the ante to sit at the table. What's going to provide the competitive advantage is how you use the data from the RFID system. Referring back to that model I talked about earlier, I would encourage companies to think about their pain points and think about the data they need to overcome those pain points. They can then think about how RFID can produce that data for them and start delivering real business value.
View All Blogs
