The food industry has been anxiously awaiting a bounty of benefits from intelligent tags for many years. Now that the first significant steps toward implementation of RFID tagging have been taken, those at the forefront of the adaptation realise that there is still a long way to go.
Elsevier Food International, Vol. 6, Number 2, May 2003
David Litwak

Imagine being able to scan the contents of an entire trailer of products while the truck pulls into the distribution centre loading dock. How about being able to track a production batch of frankfurters all the way from the processing line through the distribution centre to the store and home to the consumer's table. These are just a few of the many 'daydream' scenarios that have been envisioned for the now evolving technology known as Automatic Identification, a term that refers to the ability of an identification device or tag to be scanned, recorded and identified without active human intervention. This is being accomplished with Radio Frequency Identification (RFID) technology.
An RFID system consists of a label or tag that is embedded with a single chip computer and an antenna. The tag is similar to a bar code but is more 'intelligent' because, depending on the tag type, it can store much more information than the standard bar code. There are even RFID tags, known as Read-Write tags that can have information written to them at any point along the supply chain.
Besides the amount and type of data that these intelligent tags can hold, another major point of difference between RFID tags and bar codes is that, unlike bar codes, RFID tags do not depend on being visually read or scanned. Since the RFID system does not depend on visual scanning, the chips or tags can be embedded anywhere on or in the tagged object and more than one tag can be read at a time by the reading device.
RFID systems provide an excellent business tool that helps companies manage their supply chain better, increase their margins and profits, and decrease costs. The promise of multiple benefits from an automated identification system, usually referred to as Auto-Hi, is very inviting. However, Auto-ID and the widespread use of intelligent RFID tagging are still a long way from becoming commonplace in the food industry. The truth is that Auto-ID and intelligent tagging are still very much in their infancy in any industry.

Pilot phase
While some industries, such as the automotive and clothing industries, are ahead of the food industry in adopting this new technology, even they are still experimenting with different types of tags and readers, as well as with different standards for the entire process. The technology is still very much in the pilot test phase.

"RFID offers the opportunity to read and write to tags thus changing the data on an item without the need to replace a label or to look up data on a central database"

At the forefront of the development and testing of RFID protocols is a group of manufacturers, retailers, shippers and universities that have formed the AUTO-ID Center. Headquartered at the Massachusetts Institute of Technology outside Boston, with laboratories and research facilities also located at Cambridge University in the UK and the University of Adelaide in Australia, the AUTO-ID Center is designing, building, testing, and deploying an infrastructure for a global RFID network that would be a layer on top of the Internet. It has conducted pilot tests on RFID technology in the supply chain and is working on a set of standards for the chips and readers.
One of the activities that the centre is undertaking is the development of the Electronic Product Code (ePC). This is the next generation of product identification. The ePC is similar to the UPC bar code but has the ability to store far more data, and data that are unique to that particular product, or batch, or shipment. Besides ePC, the centre is also working on standards for the tags and the readers, the Object Name Service. This tells the computer where to locate information on the Internet about any object carrying an ePC using Physical Markup Language (a new standard for describing a physical object), and a software technology for managing and moving the data called Savant.
Starting in October 2001, the centre conducted Phase I level testing of RFID technology. The study was done on pallet level tagging using a Wal-Mart (Sam's Club) distribution centre in Texas and shipments from Procter & Gamble, Gillette and Unilever. The primary objective of this test was to evaluate the performance of the various elements of the technology. Phase II, which began last summer, focuses on testing the system's ability to handle large amounts of inventory. The AUTO-ID Center plans to conduct a third phase of testing, which will focus on testing the prototype cheap tags and recorder, taking the system to the next level of complexity.

Food industry involvement
Many suppliers and retailers in the food industry are involved in testing RFID and are among the 96 active members of the AUTO-ID Center and many more are thinking about examining the emerging technology. There are today no food industry companies that have implemented

Returnable plastic containers in use in a grocery store with an RFID tag.

RFID product identification as an everyday business procedure. Moreover. the world seems to be on equal footing in implementing the technology. Companies on both sides of the Atlantic are involved in the testing, although the bulk of the activity seems to be centred in the US and the UK. The AUTO-ID Center has recently opened a new testing facility in Japan. Aside from the tests carried out by the centre, a number of suppliers and retailers have been conducting their own tests. For instance, UK retailer Marks & Spencer has been working with lntellident Limited to use RFID tagging on pallets and returnable plastic trays containing their fresh prepared foods. According to Intellident's Dr. Andrew Dean, "The company embarked on a well-reported design and build program (using the Intellident multi-scanning unit) with Marks & Spencer, to support what is considered to be the UK's largest use of RFID tags in the supply chain management arena with over 3.5 million crates equipped with RFID tags for fresh and chilled goods, being phased in during the next two or three years."
"Why RFID? - Well, it boils down to flexibility, speed, accuracy and opportunity," says Keith Mahoney, logistics controller, food division for Marks & Spencer. "RFID offers the opportunity to read and write to tags, thus changing the data on an item without the need to replace a label or to look up data on a central database. Although we may take the central database route in the future it does not currently exist and we needed a low cost trial to prove the technology. RFID does not need line of sight, items can be read simultaneously and the technology is reusable and resilient."
"The ability to change data 'in flight' and the speed of read would enable us to move the current first read of bar codes at depots back to the point of production for our larger suppliers. This enables them to control production and dispatch more accurately, reducing wastage and extending production time through accurate advanced notification via the web," Mahoney says.
Everett, Washington-based lntermec Technologies, is conducting a similar test of RFID with reusable plastic containers. Especially for use with perishable items, such as produce, the containers can use RFID to track their contents from farm to supermarket. Intermec uses read-write tags on these containers since information on the movement of the product can be added to the tag right in the field and at each successive point in the supply chain. While the read-write tags can hold a valuable source of information, their cost and size prohibit their use on many applications.
"Every application has to stand on its own," says Mike Sitek, Intermec's business development manager for retail. "Nobody believes that one technology across the board is the solution. What's the value of writing data to a tag? That's certainly application dependent. You can associate product movement with specific product characteristics, such as batch or farm."
In another major supply chain study, Gillette has tagged nearly a half-billion packages of its Mach 3 razors and blades. The test will track the products through the supply chain of partnered retailers. One of those retailers is Tesco. According to the retailer's technology director Colin Cobain, "tagging products will lead to real improvements for both customers and staff with instant information and improved availability."
Another large pilot test is currently being conducted in the US by CHEP. The company, based in Orlando, Florida, has equipped 250,000 of its pallets with RFID tags. The company will track those pallets across six CHEP service centres in Florida to 34 customers and 2,000 distribution points. Based on the results of the test, CHEP's management will be putting together a business plan for implementation of RFID in the future.
"In the past when we were looking at technologies such as bar codes we were prevented from using it simply because of the difficulty of reading those codes", says Edwin Birnbaum, vice president of supply information systems at CHEP. "A full truckload usually consists of around 500 pallets, which take about ten minutes to load at a CHEP facility. Bar code reading would force you to scan each pallet individually at about five seconds per pallet, which would add considerable costs -hence the need for RFID."
Birnbaum says that the costs of implementing RFID have come down tremendously over the past five years. Even at about 20 cents a tag, which is far too expensive to use on individual products, the price is insignificant when used at the pallet level.
"The cost of technology is such that today it is not realistic to put an RFID on a bottle of shampoo," Birnbaum says. "A bar code can be put on essentially for free, so even a one-cent RFID is costly compared to the price of an individual product. But it may not be so in ten years because the infrastructure starts on a hierarchical basis."

Future perspective
For RFID to fulfil many of its promises beyond simple supply chain management, vendors and retailers will need to be able to manipulate the data that is flowing from the tags across the Internet. A whole data network will have to be constructed to support RFID, the standards of which are being worked on by the AUTO-ID Center and by private companies, such as Teradata, a division of NCR.
"We're making sure that we have involvement in the data standards and the ePC standards, how these evolve and how these work with our logical data model," says Dan Odette, Teradata's vice president, manufacturing industry marketing in Atlanta. "When someone builds a data warehouse they need a model of how the data looks, kind of like a blueprint for the data. We're making sure that those models are consistent with RFID data."
What is the future of RFID in the food industry?
Certainly, the momentum of the technology has reached the point where it will be hard to stop. Companies are too interested in RFID and the potential benefits not to implement it in at least the distribution chain as soon as the costs come down substantially. This should be within the next two or three years. According to a report prepared by Accenture, RFID is expected to increase sales by one to two per cent by reducing out-of-stocks, decrease inventory ten to 30 per cent by reducing the amount of minimum stock needed for immediate demand and increase manufacturing and distribution capacity and asset utilisation while reducing capital expenditure.
"Our research indicates that pallet and case-level tagging will be widespread within one or two years," says Lyle Ginsburg, managing partner for technology innovation in Accentures products operating group. "Among the drivers of this adoption are standards, lower-cost tags and retailer support. Manufacturers should prepare to meet the demands of their biggest customers, who are already - or will soon be - requiring suppliers to employ Auto-ID technology."