How does RFID land in asset management?
Many people think that RFID asset management is a three-step "buying labels, putting them on and sweeping" process, with the result that no significant increase in the efficiency of the project's up-line after-check count has been made, and often the tags have fallen off. What RFID really wants to address is the four levels behind the word "account for the same" - the layer of demand, the matching of the label, process reengineering, system integration, and the absence of one layer of carding.
Undo one misunderstanding: RFID is not just "labelled."
RFID is a way of identifying information by radiowave. The "electronic label" that comes from the reader when it comes to the chip plus antenna on an asset, the memory is returned and the reader sends the data back to the back-office system.
Common RFID tags come in three types: passive (powered by a reader), active (battery-powered and able to transmit), and semi-active. Passive UHF (860–960 MHz) is most widely used for asset counts. It reads at a distance, costs little, and can capture hundreds of tags per pass. Unlike one-at-a-time barcode scanning, items can move through a portal in batches.
And because of this, many project leaders mistakenly think that when you buy a reader and label, you do half the work. In fact, if you put it on a physical layer, if you can keep it right on the books, if you can run the business process well, that's the real workload.
First level: first, we figure out what to do and then decide how to do it.
The first step in landing is not to buy, but to split assets. A factory that stores, forklifts, pallets, and a "high-value asset that moves" like this, and a fire bolt on the wall, like an "inactive low-value asset" like an underground tube, is completely different from management thinking.
The former needs to be located, retrospective, regularly counted; the latter labels RFID, which is wasted. The general approach is to score assets at three dimensions of "value x mobile X count frequency" and divide them into A, B, C categories, only for RFID systems in category A and part B.
This step seems to be a technical exercise, but it's the easiest place for a project to flip. The level of detail in the asset list is directly determined by the subsequent selection -- the list will not reach "one size at a level" and the system will be on line with the awkwardness of "no matching asset card."
Second level: labels are not bought, they're for scenes.
Also the UHF label, which is placed on wooden boxes and on metal cabinets, can be read at multiple intervals. This is because metals reflect frequency signals, while liquids absorb signals, making labels unreadable or insufficiently readable.
So the labels on the market are very fine: general paper labels for overpack, metal resistant labels (with specially designed absorbent material) for metal surfaces, soft anti-metal tags for arc or cylinder equipment, and special scene labels that are high temperature resistant to chemical corrosion. Metal surgical devices in hospitals, metal moulds in production lines, electrical appliances outside of the house, are not the same type.
The location of the reader is also well defined. The warehouse entrance and exit are suitable for a direct antenna (which allows all labels to read), the middle of the shelf is appropriate for a full-direction antenna (which has wide coverage) and the narrow channel is used for a narrow beam antenna (which reduces error). The antenna is wrong, and the recognition rate is 20 percentage points away.
Third level: no change in process, no technology.
RFID does not bring in technological upgrades, but process re-engineering. The original asset count was "Take a watch and look for it," and now it's "Automated Collection." From Section A to Section B, the asset was filled out by a transfer order, and now it has to go through the reader, otherwise there is no current in the system.
The most resistance to process change comes from the first-line user. For example, warehouse managers are used to their original ways of doing it, and the new "knit-scan" is considered a lot more than that. In this case, training alone is not enough, and new processes must be placed in the SOP (standard operating procedures), then they are mandatory for verification by the system – for example, transfer orders without going through the RFID channel, and the financial system is directly refunded.
When Shanghai Picky Technologies, for example, goes down to the ground, it usually takes two weeks with its clients to run through new processes, find a card and go online. Because many of the so-called "system bugs" are not defined, before you standardize people's and things' actions, and technology works.
The fourth layer: data flow into the current system, and drops by rhythm.
The RFID reader is a data-collection portal, and the data collected is entered into the asset management system, and then it's connected to the ERP, the financial system, and only meaningful. If the data runs in the RFID system, the financial and physical books are two sets of accounts, the value of the project is discounted.
This addresses two core issues: interface standards and data governance. At the interface level, RFID intermediates push data to the upper level of the system through API (application interface) or database view; at the data governance level, emphasis is placed on giving each asset a "single identity" - usually using fixed asset numbers or equipment serial numbers as primary keys, with RFID ' s EPC code (electronic product codes) only as supporting indexes. The error tolerance mechanism is also designed, for example, that the same asset is read over and over again in a short period of time, and the count results allow manual review of adjustments.
At the landing rhythm, the full RFID asset management project is generally in four steps:
- Phase I (1-2 months): asset inventory + classification, output of the RFID asset list and labeling options;
- Phase II (2-3 months): implementation in pilot areas, usually by selecting a warehouse or a department of enterprise to cover procurement, warehousing, access, inventory, redeployment, end-of-life processes;
- Phase III (3-6 months): system integration + process solidification, replication of pilot experiences to other regions, and completion of interfaces between ERP and financial systems;
- Phase IV (continuing): Data governance + continuous optimization, with emphasis on cleaning up historical differences and establishing a regularized mechanism for periodic inventory counts.
The first is that the pilot is too big, the whole company is up and there's no way to end it; the second is that process retrofits are not linked to performance appraisals, and employees do not take new processes seriously; the third is that interface development cycles are underestimated, and RFID systems and ERP fields are often more time-consuming than RFID itself; and the fourth is that the physical life of the label is ignored, with the labels on outdoor assets 3-5 years old and need to be included in the annual replacement budget.
The value of RFID asset management is not in the technology itself, but in turning "unseeable assets" into "visible data." For companies that are evaluating RFID projects, it is suggested to answer three questions: What kind of assets do you want to manage? Where are your current process bottlenecks? Can your financial system digest these data? Think clearly about these three questions, and then choose and supply them, less curves.

Ms. Lei
Mr. Pang