The big question with RFID used to be, “Does it really work?” Today, the question is more, “How can I make it work for me?” That’s a clear indication of the growing maturity of RFID technologies in an ever-increasing range of applications—from identification of item-level consumer goods to oil drilling pipe—but it’s also a bit of a caution. The “how” applies as much to the business processes with which RFID will be used as it does to the technical aspects of implementing RFID.
Before we get into “lessons learned,” let’s look at some success stories. But remember, this is just a quick list with some examples to get you thinking and is not intended to be comprehensive.
Asset Identification, Tracking and Management
This is one of the hottest areas of new implementation because it has a very good ROI, particularly for moveable assets, equipment that requires periodic calibration and IT assets that may contain sensitive data.
Quickly locating moveable assets, whether in a healthcare environment or on a manufacturing floor, is essential to efficient operations. There are numerous case studies of real time locating systems (RTLS) being implemented in healthcare which not only reduced staff time looking for equipment and improved the patient experience but also, in some cases, revealed that a facility had a surplus of equipment, resulting in savings greater than the cost of the RTLS implementation. Not a bad ROI.
For equipment that requires calibration and inspection, the use of RTLS tags along with a bar code label (with different data) allows equipment to be quickly located. The technician scans the bar code label to verify the equipment has been located and, it then has to be assumed, serviced.
Safety equipment, whether moveable or in containers, can be quickly inventoried and, if maintenance/expiration data is stored on the tag, checked for readiness.
Inventorying IT assets such as removable hard drives and laptop computers has become a major issue with many companies. Passive UHF RFID tags can be used for quick inventories and to check for items inadvertently being taken out of a facility without authorization. (They’re not much good if someone is trying to steal items. In that case, a routine inventory of devices to quickly identify missing items and/or searches at exits would be required.)
Returnable Transport Items (RTIs)
In a way, this is also asset identification and management. The big difference is that these items leave your control when they’re sent out.
RFID is being successfully implemented to identify a wide range of RTIs, from gas cylinders and totes to racks and fixtures. A clear indication of the importance of RTI identification is that the major pallet rental companies are implementing RFID to track use, damage source and cleaning/sterilization.
When you hear of item management and RFID, you might immediately think of retail items. While retail is one area where RFID is being more widely deployed, particularly in apparel and footwear, you should also think of items such as oil drilling pipe, building materials, HVACR equipment and ducting, and heavy equipment tires. In other words, just about anything.
In just about any operation, you cannot operate efficiently without knowing what you have, how much you have and where it is. For building projects, ensuring that the right components are on hand (reading during receiving and laydown) means you can keep a project on schedule. Reading the item’s ID during placement ensures that the right component goes in the right place.
Item Unique Identification (IUID)
Admittedly, IUID is the term used by the U.S. Department of Defense (DoD) for its asset identification and management program. It is used here because the DoD’s IUID criteria provide an excellent basis for any asset ID system. The IUID program stipulates a unique ID for items that are:
• Over $5,000 in value
• Already serially tracked
• Mission critical
• Anything else a project manager deems should be serially tracked
In a manufacturing facility, for example, individual cutting tools can be identified and monitored for wear so that they can be used for their full expected life but be replaced before they break and ruin the piece being machined.
You might assume that the item’s own serial number should be adequate for this purpose but unless a code assignment authority and vendor code is also stipulated, the serial number itself is often not unique.
The purpose of IUID is not just asset identification and tracking. Its use for financial management is as important if not more important. An IUID pointing to a database entry with date of acquisition, cost, maintenance history and other relevant attributes allows tighter financial accountability and control.
And an IUID is at the heart of all of the preceding applications.
In most cases, lessons learned are negative examples. We often learn more when we get smacked in the head than when we get a pat on the back. It’s just the way things are.
Some may be familiar with “The 10 Commandments of Bar Codes” from a number of years ago. Here is a somewhat similar list, “The 7 Deadly Sins of RFID System Design and Implementation”:
1. Dumb Implementation
The adage that applying automation to a bad business process only gets you more useless information faster is doubly true for RFID. It refers not only to automating existing bad processes but also to bad processes developed for RFID.
We all probably understand the idea of automating existing bad processes but here’s an example of a bad process developed specifically for an RFID implementation. The company wanted to use RFID readers on lift trucks to record storage location during pick/put away operations. The vendor recommended using a photo-detector to trigger the reader when the pallet was in close proximity to the rack location. The customer insisted that that was “too complicated and just one more thing to break.” And they figured that reading every tag in every location in proximity to the intended storage location would be a good way to validate data already in their database. The system was a spectacular failure. The lift truck’s mobile computer continually crashed, overwhelmed by the tremendous volume of useless information the customer insisted on collecting. And, rather than use photo-detectors to trigger the reader, the company abandoned the project and blamed the vendor’s equipment.
So the first caveat is, “Know the limitations of the equipment you’ve selected.” The second caveat is, “Don’t let bad management decisions drive RFID system design and implementation.”
2. Assuming You Know Enough about RFID
RFID is not always predictable. Sometimes tags don’t have the read range you want; sometimes their range is too great. Companies have learned they need to shield RFID readers at dock doors to prevent them from reading RFID tags at an adjacent door. That kind of behavior can be predicted. What’s usually not expected is that a facility’s air ducts can channel RFID signals a long way and result in spurious reads.
Tag placement can also be an issue. Whereas standards groups spent years getting companies to place bar codes in specific locations, the same is not so true for RFID. A tag placed over a metal container or one containing certain liquids has a reduced range. Placing the tag over the natural void between cylindrical containers in a carton avoids this. Therefore, it’s often necessary to develop placement standards based on the carton’s contents.
It’s also necessary to get past the “non-line-of-sight” statement concerning RFID. It is usually not possible to read “captive” cartons (that is, with no visible edge) on a pallet. You need to read the tags as the pallet is built, not after.
3. Assuming You Have a “Closed System”
Most of the success stories above could be called closed systems. But it can be argued that, as RFID implementation increases, there is no such thing as a closed system. By their very nature, RFID tags are “promiscuous”—that is, they’re intended to be read by any appropriate reader. Thus, while your intent may be to read only your own tags, there may be other tags from other companies’ “closed systems” that enter your system with similar or even identical data.
It’s important, therefore, to develop coding structures based on internationally-recognized standards—ones that ensure that your tags can be clearly distinguished from any other tag. Data must include a code to identify the authority that assigns a company code (GS1, EPCglobal, D-U-N-S, DoD, etc.), the company code and then your data.
4. Ignoring Security/Encryption
Just as computer hacking and e-mail phishing have grown as the popularity of the Internet grew, so too RFID hacking is growing—sometimes just for “fun,” sometimes for profit. There are techniques and technologies that ensure tag-to-reader and reader-to-tag authentication to prevent unauthorized reading of data. There are encryption techniques to ensure that data cannot be interpreted or replicated. For any application, particularly those that become mission critical or those that are used to authenticate products, security must be considered to prevent adulteration or duplication of data—whether securing the tag data, the data transmission or the back-end system.
5. Ignorance of Standards Activities
Many of the issues facing companies in implementing RFID—from tag stands to data standards—are being addressed in various committees at an industry, national and international level. Even though it’s sometimes difficult to keep track of standards activities, ignorance of these ongoing activities (or even standards already in place) means a duplication of effort and, in some cases, conflicts with wider implementations that will require re-tooling of an RFID implementation.
6. Lack of Adequate Testing
Doing an RF survey for 48 or even 72 hours is woefully inadequate. It’s essential to test for extended periods of time to have a chance to observe transient RF interference from inside or outside the facility. Testing must also be done in situ with all the other equipment running to determine the sources, strengths and effects of RF interference.
7. Ignoring Privacy Concerns
When RFID-enabled employee IDs are used, particularly those that permit real- time tracking of employees, privacy concerns and popular misconceptions need to be addressed. Employees need to be informed of the capabilities of the tags being used, where data will be collected, additional reader locations (if applicable), and how the data will be used.
Looking at the success stories and lessons learned, two things stand out. First, anyone who has gone through a bar code implementation will recognize the successes and cautions. Second, when you really think about them, they represent nothing more than common sense and due diligence—well, except for the bad management decisions. That you can’t do anything about.
Bert Moore is the founder of IDAT Consulting & Education, a technology- and vendor-independent consulting firm that helps clients understand, evaluate and select AIDC technologies and solutions. He is also a member of MH&L’s Editorial Advisory Board. Visit www.idat.com.