The fault that keeps coming back.

The problem: every job closes, the problem doesn't

A pump loses its mechanical seal. A fitter replaces it, the job is closed, the pump runs. Eight months later it happens again, with a different fitter and a different work order. Each job, on its own, is a success. Together they are a pattern nobody has been shown, and each repair has treated the symptom while the cause, the way the pump is being run, carries on.

What the evidence says

Research for the UK's Health and Safety Executive on ageing plant says a maintenance log should be specific to each item of equipment, because it 'helps to identify costs, problem areas, and recurrent problems'. Its example is exactly this one:

Major investigations keep finding the same gap. At the Buncefield fuel depot, a tank gauge had stuck 14 times in three months; it 'was not recorded on the defects log and the Operations Manager was unaware of the frequency of failure'.^⁠[2] At the Texas City refinery, the maintenance system 'did not include trending reports that would alert maintenance planners to troublesome instruments or equipment that required frequent repair'.^⁠[3] In New South Wales, a court found a council had failed to undertake preventative maintenance on a sewer rising main after a similar failure of the same main three years earlier.^⁠[4]

Reliability engineering has a name for the discipline that fixes this: a closed-loop failure reporting system, in which every failure is reported and analysed, corrective action is taken 'to prevent recurrence', and 'the adequacy of implemented corrective actions is verified'.^⁠[8] The loop only closes if the second failure can be connected to the first.

What it costs

The seal is the cheap part. The US Department of Energy's pumping guide lists frequent seal and bearing replacement among the common signs that a pump is wrong for its duty, and warns that replacing them 'can keep a maintenance crew so busy that they overlook the system operating conditions that are causing most (or all) of the problems'.^⁠[5] The pump industry's life-cycle guidance adds that downtime can cost more than a pump's energy or maintenance, and that a plant's own failure history is better evidence than the supplier's figures.^⁠[6]

Across US discrete manufacturing, NIST found that plants relying most heavily on reactive maintenance were associated with 3.3 times the downtime and 16 times the defects of those relying on it least.^⁠[7] That is an association from a small survey, not a promise. The figures are traced and graded in What breakdowns cost.

How Mainte handles it

Recurrence is counted

Findings and service records on the same machine and component are matched, so a fault that comes back is shown as such: 'recurred 3× in 90 days', on the machine's findings, where the next person looks.

The whole history, in order

Readings, checks, findings, notes, service records, work orders, documents and photos sit on one timeline per machine, with who recorded each. The fitter's note about what they found the first time is there when the second job opens.

What the technician actually found

The completion note is kept as written, not reduced to a code. 'Faces scored, flush line part-blocked' is the line that tells the next person the seal was a symptom.

Reliability across the fleet

Analytics shows failure recurrence and reliability across machines, so the handful of assets that take most of the repair effort are visible as a list, not a feeling.

Observations from the plant itselfFixed plant connected through Mainte Edge

On instrumented plant, Mainte flags a pump performing below its twin or two instruments that no longer agree, with the readings behind each flag, so a developing problem is seen alongside its repair history.

Changing the maintenance

When the answer is a different maintenance plan, PM schedules fall due by calendar or by hour meter and arrive as real work orders, and on-time completion is tracked.

Costs where they belongCost module

A receipt is read, a person with authority confirms it, and the cost belongs to the machine, the job or the purchase order it was for, so what a problem has cost can be traced.

From a pattern to a decision

Seeing the pattern is not the same as knowing the answer. A third seal failure might mean the pump is being run dry, the wrong seal was fitted, the flush plan is wrong, or the pump is at the end of its life. What the record does is put those questions to the right person with the evidence attached, and keep their decision on the machine so the reasoning is still there next year:

• Repair again, with the reason this time is expected to be different.
• Investigate the cause, with the history, the operator's observations and the manual already in one place.
• Change the maintenance, as a schedule that produces work rather than a note in a spreadsheet.
• Replace it, on the strength of the machine's own record rather than the memory of its last bad week.