Tool control that survives a mine site

21 September 20266 min read

Tool control fails in a specific and predictable order, and it is almost never because people are dishonest. It fails because the system asks more of a busy person at four in the morning than that person can give it.

How a register drifts

1. 01A tool is bought and entered. So far so good.
2. 02It is issued against a name on a paper sheet, which is correct for about a week.
3. 03It is transferred between crews without anyone updating anything, because the sheet is in the store and the crew is not.
4. 04Its identification wears off, so nobody can match it to the register even when they want to.
5. 05It is repaired by someone who records the job against a model name rather than an asset number, and its history detaches from it.
6. 06At stock take it is either missing or, worse, counted as a different tool.

Durable identity comes first

Nothing else works until a tool can be identified. Stamped marks deform and wear, paint pen lasts until the wash bay, and adhesive labels do not survive hydraulic oil and dust. Laser engraving does, and it can carry a machine-readable code so identification is a scan rather than a squint.

Place the mark where use will not erase it and where a person can find it without dismantling anything. That sounds obvious and it is the most commonly botched part of a marking exercise.

Custody has to be recorded where it changes

A custody record made in the store is a record of the last time the tool was in the store. Transfers happen at the face, on the bench, at shift change. Unless the record can be made there, in seconds, on something the person already carries, it will be reconstructed later from memory, and reconstruction is fiction.

Rebuilding control of a fleet you have lost

1. 01Survey what is actually on site, including the tools that live in crew vehicles and personal boxes. This number will not match the register and that is the point.
2. 02Mark everything durably, with the asset number the register will use.
3. 03Rebuild the register from the survey, not from the old register.
4. 04Bring calibration and servicing current, worked off as a planned backlog rather than all at once.
5. 05Put custody recording where custody actually changes.
6. 06Report monthly, so drift is visible in weeks rather than at the next stock take.

Step one is the one operations want to skip, and it is the one that makes the rest possible. A calibration programme built on an inaccurate register produces accurate tools and inaccurate records.