BROWSE ALL ARTICLES BY TOPIC
Radio Frequency Identification Shows Promise for Food Safety
by Bill Hardgrave
Imagine pointing your smart phone at a head of lettuce in the grocery store and having the phone tell you what farm the lettuce came from and the date the produce arrived in the grocery store. What if your phone could even tell you what temperatures the lettuce was exposed to in transit?
Would you pay extra for that lettuce? You can bet I would.
This scenario could come to fruition with radio frequency identification, already used by some retailers for inventory control. One employee is able to wave a handheld device, reading the RFID tags on thousands of products and updating the inventory in a few minutes. Or fixed readers located on either side of a dock door can do the same job.
RFID is a form of auto-identification, like a bar code or a quick response code. An RFID tag contains a unique serial number, like a license tag. If you go to a department store and hold a package of underwear up to the light, you can see the tag. A “human readable” is printed on the outside of the tag, but the RFID guts are what matters.
RFID tags have been around for a while in garage door openers, key fobs, and toll passes. Those applications are just the tip of the iceberg when it comes to how RFID can be used. The next frontier is tracking food products through the supply chain from producer to consumer, helping to guarantee safety and quality.
The technology needed to track a product from the farm to the customer exists. If the final product has a unique identifier, it is theoretically possible to trace the product back all the way to every raw material that went into that product.
For that to happen, the use of RFID would have to be ubiquitous. Every entity along the supply chain would have to electronically record data—and that means everybody, from the small local farmer to the giant wholesaler. Then, all that data would have to be transmitted to a central database that doesn’t yet exist.
An RFID tag now costs about 8 cents, and a handheld reader costs a couple of thousand dollars. Is that prohibitively expensive for a small producer? You wouldn’t have to put a tag on every tomato—maybe on every box or every pallet of tomatoes. The cost is probably not prohibitive if large retailers and consumers insist. Wal-Mart, for example, requires some clothing suppliers to include RFID tags on removable tags or packaging.
Nor is a centralized database an impossible barrier. The Global Data Synchronisation Network (GDSN) enables companies around the globe to exchange standardized and synchronized supply chain data with trading partners. Data are obtained from bar codes, another form of auto ID, and standardized by a company called GS1.
Still, a lot of questions would have to be answered. What would the database look like? Who would have access? Should participation be mandated? Might consumers push for this kind of system?
Maybe—because such a system could make our food much safer and save millions of dollars for producers, distributors, and every entity along the supply chain.
Suppose a batch of cookie dough is recalled after testing shows the presence of Salmonella. Let’s say the whole batch was sold to five companies that shipped to 16 distribution centers, which in turn served 3,000 grocery stores. The cookie dough has to be pulled from refrigerators at all 3,000 stores, because there is no way to track the batch with more specificity.
A major insurance company, Hartford Financial Services Group, is encouraging transportation companies to use temperature sensors — another kind of RFID — on pallets and cases of perishable products.
But what if RFID tags were affixed to cases of cookie dough? The cookie dough manufacturer would know the precise date when the bad cookie dough was shipped and would be able to zero in on cases shipped that day. Those 3,000 stores would be narrowed to 52 stores. The manufacturer would save millions of dollars, and we wouldn’t throw away good food.
Suppose somebody actually gets sick from the cookie dough. Once investigators know where the customer bought the cookie dough, it would be possible to trace back all the way through the supply chain, through the distribution center to the manufacturer and the particular batch.
Finally, the flour in the cookie dough is determined to be the culprit. If an RFID tag were used all the way through production and distribution, the cookie dough manufacturer could determine where the flour used in that batch was processed and where the wheat was grown.
Then trace forward, determining what other products were made with the flour. If, in this same scenario, people were to start getting sick from pizza, those outbreaks would be treated today as completely separate. If the ingredients were traceable, however, you could determine that the flour from the same producer was used in both the pizza and the cookie dough. What’s more, you could determine what other products were made using the tainted flour.
With such a system, tracing backward and forward through the supply chain, we could save money, time, and lives. Today, foodborne illnesses caused by tainted food sicken 48 million people and kill 3,000 every year. If we identified tainted food more quickly and got it off the shelves, how many lives could we save?
When people started getting sick from listeriosis last year, cantaloupes were identified as the culprit, but the source of the cantaloupes couldn’t be identified immediately. When Colorado’s Jensen Farms was finally identified, there was no way to determine which batch of cantaloupes was tainted. Using RFID technology, there would have been a unique identifier on every carton shipped, the recall could have taken place sooner, and officials could have figured out exactly where the bad cantaloupes ended up.
Businesses could save money by implementing RFID technology. When Peter Pan peanut butter was recalled in 2007 because of a Salmonella problem in one of its facilities over three years’ time, the company had to pull all its peanut butter off shelves across the United States. It cost some $50 million to recall and destroy all of that product, and Peter Pan peanut butter was not on grocery shelves for several months. In fact, sales of all peanut butter were down some 25% after the recall.
So why aren’t big producers lining up to implement RFID technology? I’ve worked with a lot of them, and they’re all concerned about safety. They don’t want anyone to get sick and die, but they look at RFID the way the average person looks at buying an insurance policy. Is the cost more than the risk? Will I actually need the insurance?
Nevertheless, recalls are increasing. A food quality manager for a regional supermarket told me they receive more than 300 recalls a year—approximately one a day. And the new FSMA gives the FDA mandatory recall authority for the first time ever.
The law also seeks to shift the FDA’s approach from responding to outbreaks to preventing them, and the FDA is charged with developing a food traceability system to quickly track and trace foods as they move through the supply chain. This is going to be tough, and we are working with the Institute of Food Technologists and others to try to find traceability solutions.
The impetus for food companies to use RFID is increasing from another sector, as well. A major insurance company, Hartford Financial Services Group, is encouraging transportation companies to use temperature sensors—another kind of RFID—on pallets and cases of perishable products to lower their insurance rates. Why? Because trucking companies often bear the cost of bad product even when they are not at fault.
These temperature sensors are expensive, costing $4 or $5 per unit. They can be reused, and when affixed to a pallet or case record temperature in real time along the supply chain. Temperature can be recorded every minute, hour, or day, and in different sections of a refrigerated truck—say, near the door as opposed to far back in the interior of the truck.
Let’s say lettuce is shipped from California, and the temperature drops below 32 degrees for the first six hours on the truck. When the produce is delivered to the grocery store, it looks fine—but within a short time the lettuce will go limp.
The insurance company wants to know where in the supply chain the problem occurred. The lettuce producer says the lettuce was fine when he put it on the truck, but the retailer says it went bad in one day. This time, the problem can be traced back to the temperature during transportation. Another time, however, the supplier might have left the product on the dock, where it was exposed to high temperatures for three hours. Another time, the retailer might have left the lettuce on the dock too long after delivery.
What if the product were shrimp or fish, and the temperature near the door was five degrees higher than it should have been? How would you like to be the consumer who got the shrimp from the pallet near the door? Without RFID temperature sensors, you don’t know where along the line the product got too hot or too cold.
Until now, RFID temperature sensors utilized nonstandard technology, which made industry-wide adoption difficult because the different technologies were not interoperable. Now, GS1 has standardized the technology. Although insurance companies and the FSMA are pushing the food service industry to improve traceability, the push will ultimately come from consumers. A strong advocacy group could embrace the issue, resulting in a database created by producers, processors, and retailers. This database could be accessible to every consumer with a smart phone who wants to know where his food came from—and how safe it really is.
Dr. Bill Hardgrave, dean of the Auburn University College of Business, is founder and director of the RFID Center at the University of Arkansas. He is a member of the advisory council for the Auburn University Food Systems Initiative, which is working with the FDA to develop food safety training for food inspectors at all levels, nationally and locally.