More data won’t fix a maintenance culture that cannot act

Eric van Riet, Field Application Engineering Manager at Fluke Corporation, argues that true operational capability is proven when sites can turn signals into safe, repeatable action.

At 02:14, a condition-monitoring alert lands on the night shift. The motor on a critical pump has been drawing higher current for several hours, even though the load hasn’t changed. The trend is small but steady. A quick thermal check shows a warm spot at the termination in the MCC, hotter than the adjacent feeds.

Now the real work starts. Do you keep running to the morning shift, or isolate and investigate? What evidence is enough to justify an interruption? Who is authorised to make that call, and who carries the risk if the decision is wrong? If the site needs a permit, an arc-flash boundary, or a second set of eyes, how quickly can that happen at night?

This is where capability shows up: in the site’s ability to turn a signal into safe, accountable action, without hesitation or heroics.

More signals, more decisions

Capability in practice is proven when systems encounter the constraints of the plant floor. It is defined by repeatable decisions and actions that stand up to safety scrutiny, shift after shift, regardless of who is on the clock. Tools can generate more data than ever; that does not mean the plant can act on it.

Alarm management is a familiar example. EEMUA notes that, as alarms proliferate, the volume generated in adverse conditions can itself cause failures through operator overload. ISA-18.2 guidance puts practical numbers around the limit: roughly one to two alarms per 10 minutes per operating position is considered manageable, and more than 10 alarms in 10 minutes is treated as a flood condition.

Those benchmarks point to the same issue we see in maintenance diagnostics. When signals multiply, response quality depends on triage rules and clear ownership, backed by an escalation path that works out of hours. Without that, monitoring can slow decisions and increase risk.

Why upskilling stalls on site

When plants talk about ‘upskilling’, the focus often lands on the tool: courses completed, dashboards understood, certificates logged. That can produce people who know where to click, while still leaving the site unsure what to do when the signal points to a developing risk. The gap at a site is often authority and judgement under real constraints.

The other failure mode is concentration. A handful of specialists become the only people trusted to interpret condition signals properly, so work queues behind them. It looks efficient on paper until a key person is off shift or already stretched. BCG describes how experience shows up in the unglamorous parts of the job – troubleshooting, finding information, verifying quality and fixing rework – and notes observations where junior technicians took up to 3.5 times longer than experienced counterparts on routine tasks, with a corresponding hit to plant availability. The same dynamic applies to diagnostics: the site may have data, but capability sits in what happens next.

There is also a well-documented ‘transfer’ problem. Training has a weak return if the workplace lacks the tools and conditions that help people apply what they have learned on the job – what researchers refer to as training-transfer strategies. Without those, upskilling stays personal. It does not become operational.

That is why the next question is structural. How does a signal move through the site, from alert to safe action, without waiting for heroics?

From signal to safe action

Once a condition signal is trusted, the hard part is what happens next – moving from ‘we’ve seen it’ to 'we've acted', with a decision that is safe, documented and defensible across shifts.

The first triage typically falls to the shift supervisor or control room: is this a nuisance trend or a developing defect that cannot be left alone? From there, interpretation usually needs electrical competence close to the work. This might be an EC&I engineer, for instance, or a senior technician who can separate “this looks abnormal” from “this warrants intervention” and can say what additional checks would reduce uncertainty.

Decision authority then has to be clear: what a technician can progress immediately, such as capturing more data, inspecting, isolating and making safe; what needs supervisor or reliability sign-off; and what triggers a controlled stop or a defined deferral with conditions.

This is where a few simple strategies prevent drift. Triage criteria stop every alert becoming a debate; they force the same questions each time – criticality, rate of change, plausible failure modes and the operational window for access. Escalation triggers stop people guessing when to pull in the right authority; they make the call explicit when risk crosses a line. A decision record that survives handover stops the same signal being “rediscovered” on the next shift. It does not need to be elaborate. It needs to capture what was observed, what was checked, what was decided, who owned the decision and when it will be reviewed – ideally with the evidence attached.

Ambiguity around ownership is where good intent breaks down into drift. When decision rights and escalation paths are fuzzy, planning turns into improvisation or delay, and both increase operational and safety risk. Backlogs grow. Work is deferred without clear conditions. The same alerts resurface because nobody is accountable for closing the loop.

A recent Scientific Reports study in a fertiliser-manufacturing unit shows the upside of treating execution discipline as part of maintenance performance: after introducing a structured evaluation scheme tied to core metrics, MTTR fell from 8.2 to 5.7 hours and MTBF rose from 27.5 to 34.4 days. The detail matters less than the lesson: when responsibility and decision gates are designed into the work, reliability outcomes move.

But what is the best way to build that competence and consistency without adding risk or swallowing the week in training?

Practice beats programmes

Capability strengthens fastest when learning sits inside the work, with a supervisor or senior engineer teaching the moves that matter. That is the logic behind Training Within Industry (TWI): a structured way to teach jobs consistently through demonstration and coached practice. NIST notes that organisations using TWI have cut the time needed to bring new hires up to productivity, alongside reductions in scrap and rework.

For condition monitoring and electrical diagnostics, the equivalent is practical. Start with a small set of scenarios the site sees repeatedly: a current trend on a critical motor, a thermal anomaly at a termination, or a partial discharge alert that needs corroboration. Build a short “first response” routine that guides what evidence to capture and when to involve the next level of authority. Then teach it through supervised repetitions on real equipment, with explicit sign-off when the risk is higher.

This is also where judgement can be made teachable: examples of good calls, what was checked, what was deferred and why, and what would have changed the decision. Continuity matters as much as competence. HSE’s human factors guidance makes the point bluntly: shift handover exists to communicate task-relevant information reliably so safe, effective work continues; failures at handover have contributed to major incidents, often during planned maintenance. A handover note that captures live degradations, deferrals and their conditions, the evidence gathered, and the next review point stops the same issue being rediscovered and re-argued.

What capability looks like in practice

At 02:14, you don’t need another dashboard. You need a decision that will still look sensible in daylight. If the specialist is off shift, the question is simple: can your site make the call safely, document it clearly, and hand it over without the story changing? Or does the alert sit in limbo until the right name is available?

That is the standard worth building towards. Operational capability is what a site can do, reliably, when the pressure is on and the next step is theirs.