Readers might remember my previous article on counting methods and RFID where I promised to revisit this topic. Well, here we are.
Interest in using RFID for warehouse physical inventory management continues to be high. The technology is very attractive and it is not difficult to wonder why it has taken so long to adopt RFID at this level. Let’s explore the challenges:
- Location, location, location. This remains an elusive problem. Using RFID to locate items down to the bin level is difficult, not necessarily because of technology limitations but mostly due to the laws of physics. There are schemes to overcome these issues, such as smart bins, triangulation, etc. but these don’t scale well and are usually cost-prohibitive. See Bouet & Dos Santos (2008) for further reading.
- Business uses. In a warehouse, we perform just a handful of basic operations: receiving, shipping, putting items away, picking items, and counting items. RFID shines when items are moving, such as cars through a toll plaza, passengers at an airport, etc. Therefore, it is no surprise that RFID does well in shipping and receiving operations at distribution centers, under proper conditions. But this is what we call “pallet level” tracking. In this article we are talking about bin-level tracking of individual product units in the warehouse. Picking and putting material away already require personnel to physically contact the items, so RFID does not solve any business problem there. That leaves physical inventory as the only potential area of opportunity, and making a large investment in RFID to improve one transaction area may not make the best business sense. It is difficult to find a paper in this area that does not sound like a sales pitch, but for further reading give Hackenbroich, G., Bornhövd, C., Haller, S., & Schaper, J. (2006) a try.
- Technical. I cannot ignore technical challenges in this article, but let me concentrate on what I call business-technical decisions. For example, the data that is associated with each tag, interfacing systems, tag design strategy, integration of technology with business practices and rules. I would argue that this is the most critical part of any RFID implementation project. Those technical decisions can have a drastic impact on business and operations. There are many papers written on this subject, but many are full of bias toward specific vendors or technologies, so a safe source is the RFID Wikipedia page.
- Workload impact. After the business-technical decisions are made, we need to consider impact on workload and productivity. In order for RFID to work for us, we would need every single unit to be tagged. Tags have to be printed/recorded, the data associated with each individual tag needs to be entered in a system, and the tag has to be physically affixed to each individual item. Let me stress that we do not mean that, if we have 10 units of a product, we would generate 10 duplicate tags – no, it is much worse, we would generate an individual tag for each unit. In other words, each tag needs to be unique if we hope to use RFID for counting. This is where some people bring up the fact that many tags have a “license plate” that already makes them unique – what that argument ignores is that each tag identifier still needs to be associated to each item being tagged. In contrast, this is never the case with barcode labels so, no matter how we slice it, we would be doing more work. Do we want our receiving process to take all that additional time and effort? Good time for a business case analysis. Does the benefit outweigh the impact? I had mentioned Daniel Hellström in one of my earlier articles, and he has a good case study that applies to this topic, see Hellström, D., & Wiberg, M. (2010). Perhaps one of the most prolific authors on the subject is Gary Gaulker; see his paper from 2010 for a good example of his work related to this topic.
- Regulations. As you might have guessed, my belief is that the principal benefit of RFID in retail warehouse operations is in physical inventory support, albeit with the known challenges about location and up-front workload. However, regulations might not even allow us to do that. Current DoD rules are not specific enough, though that may be about to change, on what constitutes a “physical count” but some auditing standards an accounting policy make it more obvious that it means human observation. So, for physical inventories that support some regulatory compliance, and unless we are talking about very short range RFID, it appears that RFID is not an option for inventory accountability until the rules specifically allow it.
This article is only a superficial discussion of the many issues facing RFID adoption for DoD warehouse physical inventory management. There are likely more areas that can be included. As usual, the issues are not all technological. Much more analysis needs to be conducted on the potential for business re-conceptualization in DoD enabled by RFID technology; until then, RFID will remain a technological solution looking for a business problem.
References
Bouet, M., & Dos Santos, A. L. (2008, November). RFID tags: Positioning principles and localization techniques. In Wireless Days, 2008. WD’08. 1st IFIP (pp. 1-5). IEEE.
Gaukler, G. M. (2011). Item-level RFID in a retail supply chain with stock-out-based substitution. IEEE Transactions on Industrial Informatics, 7(2), 362-370.
Hackenbroich, G., Bornhövd, C., Haller, S., & Schaper, J. (2006). Optimizing business processes by automatic data acquisition: RFID technology and beyond. In Ubiquitous and pervasive commerce (pp. 33-51). Springer, London.
Hellström, D., & Wiberg, M. (2010). Improving inventory accuracy using RFID technology: a case study. Assembly Automation, 30(4), 345-351.