RFID is Changing Retail

Dr. Gerd Wolfram
Dr. Gerd Wolfram

What can retailers expect from RFID ?

Wolfram: The application of innovative technologies is one of the huge trends that will play a decisive factor in retail competition in the future. RFID is a technology that enables special labels to be read via radio signal quickly and without contact over short distances of up to one meter. As more information fits on an RFID label than on a conventional barcodes, goods flows can be controlled without gaps, and product information such as the price, manufacturer, expiration date, and weight can be exchanged. There are advantages to be had in warehousing, quality assurance, and the management of merchandise categories. We can see what has been sold, withdrawn from stock, or put in the wrong place. This can help to reduce out-of-stocks. RFID promises among other things better protection against theft and guarantee processing right up to protection against plagiarism.

That sounds wonderful. How is it going to be achieved?

Wolfram: We are proceeding step by step. Up to now, we have been testing RFID as part of our Future Store Initiative within individual areas of the supply chain. In warehouse management, we are using it to automate the goods receipt control, for example. Goods deliveries to the Future Store are given RFID labels, known as “tags”, in the central warehouse, which are captured when the goods arrive in the store and when they are transported from the store warehouse to the sales area. One challenge at the moment is how to get the tags on to the product units that are put into the shelves. For this to be possible, they will have to become cheaper. It is not viable to put an RFID chip costing 40 cents onto a yogurt carton that costs 30 cents. We need to get down to a price of around five cents.

Doesn’t the different size of the goods pose a problem too?

Wolfram: It is difficult to put RFID tags on small products such as lipsticks and other cosmetics. In addition, the tags do not work perfectly under certain conditions. In the metallic environment, in particular, there are problems reading RFID chips because faults are caused by interference. Further research is being done into these technologies. We expect to have different types of transponder labels, right through to chips the size of a grain of rice, which in future it will be possible to integrate in different environments, for example in the glass of a bottle.

In connection with RFID, standardization committees have agreed to a 96-bit electronic product code – EPC – for product identification. Won’t this generate almost unmanageable expanses of data?

Wolfram: It is true that, in theory, you could classify every hydrogen molecule on the earth with such a bit length. If the complete 96 bit field was read out during each read operation in the supply chain, we would have data volumes that are too large for today’s data warehouses. We must therefore set up rules for how processes are to run. When we have the process-based rules, we simply need to filter out the data for exception messages, such as “Shelf empty” or “Delivery dispatched too late”.

Do you have software solutions for RFID-supported business processes?

Wolfram: One of the challenges we are faced with is how to link up RFID or EPC data with the operational ERP systems, SCM, and merchandise management in real time. We are actively working towards this with SAP. As a founding member of the Future Store Initiative, SAP presented a standard software for RFID a few weeks ago. This is important, because we need stable infrastructures for standard solutions, not for individual solutions, so that RFID data can be integrated into a broad range of business applications and processes. With improved data transparency and data accuracy, supply chains become adaptive, which enables us to react more flexibly to market changes.

How is standardization achieved?

Wolfram: The Metro Group alone has more than one thousand suppliers, with whom we exchange goods and information. It is of no advantage to the retail industry if a vendor requires different supply chain processes for exchanges with Metro and Wal Mart or Rewe, for example. Both retailers and consumer goods manufacturers are therefore working towards RFID standards. This started four years ago with the Auto-ID Center at MIT in Boston, and is now being carried out as an international initiative, EPC Global, involving the two large standards organizations EAN and UCC. And we are actively working as board members and in different working groups. Together with our suppliers, we also want to strengthen the collaboration with German standardization organizations such as CCG.

Up to now, the development process for standards in retailing and the consumer goods industry has been the other way round: first, the competing solutions had to prove themselves; then the industry standards were set. Are you not hoping for too much?

Wolfram: In the past, technologies have been introduced in retailing that often failed when put to the test. Today, we therefore invite manufacturers to test technologies in a concrete and real test environment that meets practical requirements. We are convinced that this is the right way to get reasonable solutions.

What are your plans for 2004?

Wolfram: In a number of steps, starting November 2004, we and around one hundred Metro suppliers will put RFID tags on all pallets and transport packaging in the various production sites for 10 central warehouses of the METRO Group. One hundred Real and Extra stores, 122 Galeria Kaufhof departments stores, and 59 Metro Cash & Carry outlets in Germany will then receive RFID-tagged deliveries from these warehouses. In order to ensure smooth implementation, we are setting up a test laboratory for the retail partners involved. Here, the functionality of the RFID technology, such as the reading of the tags, can be tested in advance. We must do a lot of convincing and provide a great deal of training, particularly for the small and midsize companies, which need simple solutions.

And when do you expect to see RFID at product level?

Wolfram: We assume that over the next three to five years, RFID will just be used on logistics units, containers, or pallets. From the present point of view, the implementing at product level will not be realised by 10 to 15 years. But high-value products such as DVDs, jewelry, or cell phones will undoubtedly have RFID labels before that.