An “advisor for nutrition and intelligent shopping?” How did that idea develop?
Hellenschmidt: BERNIE is a prototype developed as part of a Frauenhofer initiative for intelligent products and environments. The goal of the initiative is to research and develop technologies that help people with the tasks of daily life. BERNIE focuses on health. As part of the initiative, our department hit on the idea of making supermarket shopping – an everyday activity – simpler in terms of health. We wanted to create an intelligent shopping cart that would advise customers while they shop.
What technical equipment does that kind of shopping cart need?
Hellenschmidt: We absolutely insist on a system that requires as little action as possible from the user. After all, technology should simplify the task, not complicate it. Entering a barcode or handling a scanner were both excluded from the outset. Instead, we wanted the technology to provide direct assistance – while a customer filled the cart with goods. For that, we needed a technology that worked without contact and had a specific range. RFID was the obvious choice. We’ve outfitted a shopping cart with the required electronics – an RFID reader and RFID antenna – and outfitted the goods with passive RFID tags.
What kind of transmitter is involved?
Hellenschmidt: We use passive RFID tags on the goods. They aren’t transmitters in the usual sense of the word. When a passive tag is activated externally by an antenna, it oscillates its information as an answer. BERNIE uses UHF from 865 to 869 MHz. The required software is programmed in Java and can run adequately on various devices, such as personal digital assistants (PDAs) or mobile telephones. In principle, every RFID tag has a unique ID and a specific amount of memory – usually 96 bits – that can be populated as desired.
What information does the customer receive?
Hellenschmidt: In this case, we have populated the memory with a coded selection of the ingredients of various grocery items. Concretely, the ingredients cover alcohol, wheat, milk, milk protein, and meat. The prototype thus supports recovering alcoholics, those allergic to gluten or lactose, and vegetarians.
At CeBIT, you mounted a laptop on the shopping cart. What would the solution look like in a supermarket?
Hellenschmidt: We use a traditional antenna with a frequency of about 865 Mhz as the receiver. The antennas are available in several forms. At CeBIT, we decided on a bulky solution so we could clearly demonstrate the structure of the experiment to the public and get right into technical discussions. The finished product would have an invisible antenna and be invisibly integrated into the shopping cart. The laptop could easily be replaced with a small display on the handle of the cart.
What target groups does the project seek to address?
Hellenschmidt: This technology is helpful for people who need to pay attention to the ingredients of the food they consume – for reasons of health, religion, or ideology. Ultimately, we’re looking at the end consumer, but the actual technical realization of BERNIE involves upstream target groups: manufacturers of food products and grocery stores chains.
What’s the current status of research and development?
Hellenschmidt: Most of the research has already been done. Now we need to work on miniaturization and also look at user evaluations and market penetration. For example, about 20 percent of the population is lactose-intolerant and one of every 500 people cannot consume gluten. These people will benefit from BERNIE. The more these kinds of figures are known, the clearer the benefits of BERNIE will become and the more specific the information that we place on the tags will become. At the same time, we must also raise awareness that RFID technology still triggers unnecessary stomach pain in many people.
What kind of concerns do you run into?
Hellenschmidt: They are usually unspecific worries. “I don’t want to be followed,” is a typical statement that I’ve heard a lot. Here’s another: “My neighbor will know what I have in the refrigerator.” A study undertaken at Humboldt University in Berlin has looked at these concerns. The study mentions “worry about a loss of control” and “traceability” among other concerns.
How do you respond?
Hellenschmidt: First of all, we must explain the technology in more detail. An RFID tag has a very limited range even when there’s nothing but air between the antenna and the tag. A wall is enough to block the information. And we have to highlight the concrete benefits. People should recognize the added value they receive from BERNIE. Hazelnuts in a breakfast cereal put people with allergies in a life-threatening situation. Which candies contain alcohol and which do not is important information for a recovering alcoholic. Lactose, gluten, and fish are just some of the many things to which people have an allergic reaction. BERNIE could offer them concrete help in making decision while shopping – and make life a little bit easier for them.
What did you bring back from CeBIT?
Hellenschmidt: At CeBIT, we received a lot of encouragement from “normal” people – those without a lot of technical background. Many were enthusiastic about the idea and its implementation. I saw myself placed in the position of justifying why BERNIE is not yet on the market.
And now what?
Hellenschmidt: Initial conversations with representatives of supermarket chains were held at CeBIT. These conversations must be continued. When they develop positively, we’ll test the BERNIE prototypes in a few supermarkets. Each shopping cart will cost a supermarket about 1,500 Euros. Populating the RFID tags with the descriptive information and placing them on the groceries will also involve additional cost.
Let’s assume that a product is coded incorrectly and provides incorrect information. A customer buys the product and has an allergic reaction. How do you view the problem of liability?
Hellenschmidt: I’m no lawyer, but it’s precisely the wide use of such technology that can hinder such a problem. As a representative of the food producer, the supermarket would guarantee that product would not harm the consumer. Legal questions would probably develop in the precise relationship between the food manufacturer and the supermarket.
Among the suppliers, who decides on using BERNIE – the supermarket or the manufacturer who puts the tags on the packaging?
Hellenschmidt: The supermarket would have to equip the shopping carts and provide the software. All manufacturers would have to agree to a mandatory and uniform description of the product, including information on its ingredients. A legislative push would accelerate things. Theoretically, BERNIE could be implemented in 6 to 12 months. In practical terms, industry, retail, and legislatures would have to reach an agreement, but experience teaches us that hurdles will arise, so actual implementation would have to wait for a considerable time.
What advantages would a supermarket have in using the technology?
Hellenschmidt: A supermarket that decided to offer this assistance to its customers would of course appear as an entrepreneur. That can be a tremendous commercial advantage – particularly in light of the increasing consciousness of many consumers about their nutrition.
How do you evaluate the significance of RFID technology? What are its strengths and weaknesses?
Hellenschmidt: The benefit clearly lies in the lack of contact when reading information about goods. Unlike barcodes and other technologies, RFID enables reading more information simultaneously without additional manual intervention. That’s immeasurable for logistics and “intelligence” in the supply chain.
I see weaknesses in two areas. The first is the cost factor. Even passive tags currently cost a few cents. Active tags are even more expensive because of their own power supply from batteries. The second problem is technical. The range of passive tags is only a few meters. The shielding effect of water or glass is also a handicap that comes from the physical properties of alternating electromagnetic fields, which are the foundation of RFID technology. The technological problems will be resolved in the near future. And even today, deft settings – like an increased number of RFID antennae or placing more than one RFID tag on an item – make such problems a thing of the past. Of course, these items would initially increase costs, but the costs can be limited with mass production or new manufacturing technologies. There are already attempts to print RFID tags.