nfc tag inspection programme: from patrol point to data closed
RFID Knowledge

nfc tag inspection programme: from patrol point to data closed

PublishedShanghai Zejie Smart Technology Co., Ltd.

NFC inspection is not dependent on tags.

NFC labels are checked, and the core is to turn each of these more probabilities into a verifiable, traceable data node. The key is not to run through, but to plan for the tag itself, while the collection links and backstage handle these three places. The applicable condition is: indoor or semi-indoor, point-deep, need to prevent the carding scene. We're trying repeatedly in factories, industries, distribution rooms, several types of projects, and we're sinking out a relatively sound set of practices, which are then broken down by steps.

Let's decide if we need to use NFC.

NFC is not a gold oil. Take the scene apart and answer at least three questions: whether the inspection area is mainly indoor or outdoor? Does the inspector have a single distribution cell phone? Does the patrol need to prevent a call card?

The advantages of NFC are concentrated on three points, compared to a few common scenarios. First, physical uniqueness, the NFC label has chip-out UID (the only global number), normal QR code is easily copied and can be replicated, while the NFC tag UID and writing together are tested with much greater difficulty in taking the form. Second, there is no power, the label itself does not use batteries, and it is attached to the device for five to eight years. Third, close sensing, reading card distances usually 2 to 5 centimetres must be applied in close proximity, which naturally limits remote cheating space.

In turn, NFC has an inappropriate scene. The Minister of Foreign Affairs is a wide range of miles from the line (oil and gas pipeline, forest fire protection), and several hundred metres from point to point, with slow and hard-forward sensing by the inspector, and GPS patrols more appropriate. Pure outdoor open areas, NFC labels are also vulnerable to suntan and rain, and the selection must be treated in waterproof conditions.

The experience has been drawn: the interior is heavily equipped, with NFC being given priority for the scenes required to protect against theft; the pure Foreign Minister of Taiwan is on the line first and GPS is on the line; budget constraints and tolerance for a certain number of secondary checkers are enough.

The more we're planning, the better.

The spot planning is the skeleton of the inspection programme, which was wrong and cannot be saved from hardware or software.

The first principle is to set a point by the motion line, rather than by the device. First, draw the inspector's actual walk and then place it on the key node of the action line (equipment side, floor corner, risk source). We did a warehouse project that initially planned over 40 NFC points at "one point per device", and inspectors walked for two hours in a shift, with a high leakage rate. Then we changed to "line-by-line node planning," and then we reduced the spot to 18, and the leak rate was half.

The exact number of units can be used to refer to these empirical values. The single floors, around 2000 square metres, are controlled by 8 to 12 points; the small room, which is a distribution room, is 4 to 6 per floor; and the one floor of the writing building is 6 to 10. The walking time between adjacent locations is suggested at 3 to 8 minutes, too close to make it redundant, too far to create a "neither go" sluggy.

And you can avoid a few types of pits. The dots are not packed behind the door, the equipment is stitched, the pipe is covered, and inspectors are able to find them within 30 seconds. High-level advice is 1.2 to 1.5 metres, which corresponds to the position of adults who raise their hands. The high-altitude dots (over 2 metres) are also used for longer poles, but it is important to consider whether the inspector is willing to pull out his tools at each time.

NFC Tag Selection and Deployment Details

The selection phase has a lot of pits that can easily be stepped on.

For label chips, NTAG213, NTAG 215, NTAG216 are the main models. 213 capacity 144 bytes, sufficient to store point number and a small amount of information; 215 and 216 are larger, slightly higher prices, suitable for complex scenarios that need to be written into equipment parameters, as shown by the inspection. The normal inspection 213 is enough, there is no need to follow up on specifications.

The outer-hull material directly determines the life of the body. The indoor drying environment is just a normal PVC sticker or ABS shell, with a unit price of a few cents. The outdoor or wet environment (the ground-debreave, the coastal distribution room) must select IP68 protection level, an epoxy-enriched metal or hard plastic case, otherwise it will be degelized in two or three months, and read card distance decline. We suffered losses at an early stage in a coastal distribution room project, and the general sticker will have a defouling rate of over 30% after three months.

Metal surfaces are another common snag. NFC communicates by electromagnetic induction, so attaching a tag to a metal enclosure or pipe can absorb the signal and reduce its read distance from 5 cm to zero. Use an on-metal tag with a backing layer or add at least 3 mm of high-frequency RF-absorbing material. If the budget allows, buying finished tags with a built-in on-metal layer is the simplest option.

Do not write just a number to the tag. A suggested format is checkpoint ID, verification code, and a short description, such as "PT-A03-C8D9 / Inspect Main Transformer No. 1 Fan." The verification code helps prevent counterfeiting: generate four random characters when writing the tag and check them when reading it, so someone cannot simply buy a blank tag and impersonate a checkpoint.

The link between the card and the data repository.

Data links seem to be an IT department, but the most on-site step is this.

There are usually two reader options: inspectors' own NFC phones or industrial handhelds provided for the project. Phones cost less and are easier to maintain, but check that they support NFC, which is omitted on many budget Android models, and that it is enabled by default. Industrial handhelds cost several hundred to over 1,000 yuan each, read reliably, and can withstand water and drops, making them suitable for outdoor or high-frequency use.

The most problematic thing is the offline cache. Ground-debreave, elevator wells, people's defense works in these areas, mobile phones often fail to signal, reading card data can be lost directly if local caches are not available. The norm is that cell-side caches have recently been recorded 500 dial cards, and the network automatically uploads them in chronological order once they are restored. It is recommended that each time a card is called, do not save for batch reports -- the side effect of the backlash is that once the network is lost before it is uploaded, the whole series may be thrown out because of a spill.

Time tags are also noted. Cell phones can be manually modified by the inspector, and if local time is used to make a card determination, the inspector can change the "call card" in advance. It is recommended that the backstage should be scheduled for the NTP server (over 2 minutes of alarm error), or that the backstage should use the server to overwhelm local time when it receives the call card data.

Another small detail: NFC label reads with the risk of rereading. The inspector may read it several times during his stay near a point, creating double-clicked card data. Read the card end to weigh it down, and the rule can be "One time in 60 seconds for the same tab."

Let the data really be used.

Data collection is only the beginning, and it is the subsequent paragraph that determines the value of the programme.

First define what counts as an exception. Four common types are overdue patrols (a checkpoint is not reached in its time window), missed checkpoints (a required point is not read), route deviations (the actual check-in sequence strays too far from the planned one), and too many retries after a read failure (more than five at the same point). Once rules are set, automatically assign each exception to the responsible person, such as the equipment supervisor, with a response deadline. Four hours is recommended.

The second thing is the board and the rediscovery. The class viewer shows a completion rate for the day, an anomaly point Top3, a list of leakers. Project-level panels look at trends by week and month, identifying "systemic leakage areas" and "stability and reliability points." Monthal revisits are easily ignored -- aggregate monthly anomalies, looking back at which points frequently call the police. If one of the points keeps calling the police within three months, but no real problems are detected in the on-site inspection, it is probably that the dot itself is not well planned (overspectrum, severe cover, unfeasible inspector) and should be adjusted or removed directly.

Data retention cycle recommends no less than 12 months. The value of the inspection data is not only a current recourse, but also a basis for ex post-retroactivity -- equipment malfunctioning, calling the inspection records over the last three months to see if there are any omissions, and whether there are any premonitions.

When doing such projects, Shanghai is accustomed to do site planning, rule configuration, board-watching templates as part of standard deliverables, and avoid typographically searching for each project. If you are evaluating the NFC inspection program, it is recommended that the point-planning and anomalies rules be clearly defined, and hardware selection and backstage development can go in parallel, so that the overall cycle can be reduced by two to three weeks.