If you and I had a conversation about radio frequency identification (RFID), we might talk about how you wanted to implement it, or maybe you’d tell me about the headaches and frustrations you had when you installed it at a facility. I would share that the technology is not what causes the struggles. The problems come in applying the technology.
RFID is important in manufacturing. It helps track vehicle loads, provides guidelines and boundaries to ensure product safety, and streamlines processes to require less hands-on management. These benefits are easier to achieve if you’ve already thought through technological limitations. Radio frequencies are part technology and part form, i.e., you need to find a sweet spot to catch as much signal as you can with a minimum of fuss.
Based on our experience as a system integrator implementing custom vehicle tagging applications using high frequency (HF) RFID, it’s important for the customer to explain the level of performance they expect. As an end user, some of the questions you should ask a system integrator about an RFID project include:
- How long do you think it will take for this to work well at this facility?
- What improvement in process do you expect?
- How much will it cost to get that improvement?
- What read percentage are you expecting?
It isn’t possible to get 100 percent correct reading of RFID tags without controlling a lot of variables for where tags are placed, how they are read, and where to install the antenna so it works best, regardless of how it looks. Few sites are able, let alone willing, to invest the time and money necessary to control everything for the sake of achieving perfect read percentage. Therefore, expectations need to be set early in the process.
Only after defining your performance expectations can you make decisions regarding the technology to implement. There are many thousands of types of tags and multiple sources for readers. If your need is for technology like that is used for vehicle tags on a toll road, your tag and reader choices are quite different than if you are looking for solutions in truck receiving areas where you are depending upon drivers to place tags where they can be easily scanned by a reader. Once you choose the tag and reader, you’ll have to decide how much control to have over where the tag is placed. It is not a guarantee that a tag will only ever appear in one place, as that isn’t always possible.
Before RFID implementation begins, clients must consider how they will handle circumstances that aren’t ideal. For example, a tag might hang over one truck’s door handle and work perfectly. However, the next truck’s door handle is out of the read area and will not work. A tag may be in the right place on a truck’s windshield, but the windshield is curved and distorts what the reader can find; or perhaps the tag is clear, but the reader’s position is blocked by a side mirror on a particular make of truck. While the technology is sound, these details that are unique to every truck delivering to a facility make it difficult to implement. There is investment on both sides of the tag, and drivers must benefit in some way from participating in the RFID process; otherwise, the system will not provide the expected results.
RFID implementations will never be boilerplate; every site is a process of fine tuning for a specific facility. There are considerations with where to affix the tags, where to place the reader, where to mount the antenna, and fine tuning the signal so the reader “sees” the tags we want it to. It can be a difficult thing to estimate the time and resources needed to correctly place and mount the technology selected when implementing a custom solution. Some estimating ability comes with experience, but we also want to help clients understand the limitations of RFID and match those up with their expectations before embarking on this path.
I firmly believe the technology itself is reliable and adds value. A smooth transition to daily use at a facility requires asking the right questions about expectations for the technology, timelines and percentage of correct reads. These are the points that should define RFID implementation.
Ben Langstraat is lead control systems analyst at Interstates Control Systems Inc. a certified member of the Control System Integrators Association. See Interstates’ profile on the Industrial Automation Exchange by CSIA.