Small-scale RFID systems can achieve big benefits
A highly selective approach to RFID tagging can result in great return on investment (ROI). It is widely understood that tagging can be limited to high-value items, where tag cost is small in comparison to the item cost, preferably for applications where the item is used over and over again. However, it is also possible and sometimes very compelling to tag a small percentage of relatively inexpensive items.
When people look at RFID technologies the first question is inevitably, "How much do the tags cost?" Regardless of the answer to that question-be it $50, $5, $1, or 50 cents-potential users are inevitably disappointed. Industry visionaries predict a day when billions of tags will cost 5 cents each. Some people assert that the 5-cent tag is possible today, if only somebody would place an order for billions of tags.
Even at pennies per tag, it's hard to get away from arithmetic realities. You can't put tags on everything. Five cents multiplied by "everything" is a very big number.
For a new application, we've found that a smaller system and realistic short-term objectives can be successful by following these steps:
Step 1-Process sampling: Start by tagging a sample of items (5 to 10%, or less) to measure business processes. An RFID sampling system may be used on a temporary or periodic basis to realize these benefits.
Step 2-Process monitoring: The first ongoing application may be as simple as running the sampling system on a continuous basis. Such a system has an internal justification that may not shift paradigms, but has a good ROI and gets the system in place.
Step 3-Incremental system expansion: Layer on tags, process improvements, software and partners-in self-justifying baby steps-over time to get the full benefits. The tagging approach and software architecture chosen in early phases should enable such expansion.
We have been using this design approach in our consulting practice, and have found that it helps to break through the objections to using RFID systems. The question changes from, "What does the tag cost?" to "How does improved data justify the total system cost?" As a rule of thumb, we've found that as long as the proportion of tagged items stays in the range of 5 to 10%, the cost of the tag becomes an important, but manageable, consideration.
CHEP starts small with Smart Pallets
An excellent example of the sampling approach is in the pallet tracking application by CHEP International, Orlando, Fla., an international pallet and container pooling company. CHEP's "Smart Pallet" initiative was one of the most hotly contested applications of 2001, with more than 40 hardware and software vendors being evaluated.
Presentations and discussions of the CHEP Smart Pallet paint a future of almost magical tracking of goods through the supply chain. The notion is that once you've tagged the pallet, you've also tagged the goods on the pallet.
But CHEP plans to tag only about 10% of its 200 million pallets and containers on a randomized basis. If only a random 10% of pallets are tagged, how can they use the tags to track specific orders? Ten percent is only helpful if your goal is to track a population of assets on a statistical basis-and that's exactly what CHEP has in mind for its initial system.
CHEP justifies the Smart Pallet RFID initiative purely on the basis of better management of its large inventory of pallets. CHEP's basic business is to deliver a large number of essentially identical standard pallets to a supplier, and pick up the pallets later at their ultimate end-user destinations, such as a Wal-Mart store.
Pallets at a particular end-user location typically come from multiple suppliers; and there is currently no way for CHEP to determine where a pallet originally came from. If a pallet is damaged, who is responsible? If pallets delivered to a particular supplier or end-user tend to "disappear," there is no way to determine that this is happening. Since all pallets look identical, CHEP only knows that a certain number of pallets of a particular type have been dropped off or picked up from particular locations. Making a connection between those two events is basically guesswork.
By placing an RFID tag on a relatively small percentage of its pallets, CHEP gains visibility into its asset base. CHEP engaged mathematicians to find the optimal percentage of tagged pallets. The brains recommended tagging 10% of the pallets on a random basis to provide the desired asset visibility.
Tags ID supply chain snags
With CHEP's new system, Smart Pallets are archived when an order of pallets is dispatched to a supplier. When the Smart Pallet re-appears at a CHEP facility, the time and place of return is noted, along with the identity of the end-user. If a Smart Pallet is damaged, the damage can be linked back to the originating customer and the pickup location. If a Smart Pallet is lost, the loss may be attributed to a particular customer. Thus, by tagging only a small percentage of tags and only reading them in CHEP's own facilities, CHEP will build a reliable statistical picture of where problems are occurring, giving the company an opportunity to work with customers to resolve these problems.
Andy Robson, business development manager at CHEP, enumerates the following potential benefits from RFID tagging:
· - Help in further identifying misuse and breakage by supply chain users
· - Improved cost visibility by customer (breakage, dwell time, losses)
· - More accurate picture on the number of pallet losses
· - More timely and accurate pallet reconciliation
· - Reduced conflict with customers
· - Reduced customer administration
· - Value-adding opportunities for product-tracking applications
CHEP is in the process of running a limited pilot of 250,000 Smart Pallets. The main question is not whether the system will technically work, but whether CHEP can operationalize the data from this technology to get measurable improvements.
CHEP and its development partners have been aggressively promoting the value-adding opportunities to link to product-tracking applications, with the Smart Pallet as the key to supply chain visibility.
This is indeed an interesting vision, but how do we get from here to there? We see this happening in baby steps. First, some CHEP customers will take advantage of the RFID tags that come "free" on some 10% of their pallets, and use these tags to learn more about their own processes on a sampling basis. This early experience, if successful, will provide basic confidence in the technology, opening the door to using the pallets to track specific goods through the supply chain. Customers will begin to ask for Smart Pallets for certain applications, which CHEP will of course happily provide, perhaps for an extra fee.
As the use of Smart Pallets for supply chain management extends, and the price of RFID decreases, eventually it may become less expensive to simply build RFID capability into every pallet, rather than maintain separate inventories of "smart" and "dumb" pallets.
The general lesson is that it's good to have a vision of where your RFID application can go in the future, but it's even more important to define short-term objectives that are achievable and have compelling business results. Process measurement is often a good place to start.
