Condition-based maintenance cost: what it costs to listen to your equipment

A vibration analyst walks the floor with a data collector, touching sensor points on pumps and motors. An oil sampler draws fluid from gearboxes on a set route. A thermographer sweeps a camera across the switchgear. Not one of them repairs a thing today. They are listening, for the faint early signals of trouble that a calendar-based schedule would sail straight past.

Condition-based maintenance cost is what that listening costs, measured as a share of the maintenance budget. It looks like pure overhead until you notice what it does to everything else: a healthy spend here is very often the reason the reactive repair bill starts to fall.

What it actually measures

CBM cost is the percentage of your total maintenance spend that goes on measuring, trending and comparing equipment condition against known standards, so problems are caught before they become failures. It covers vibration analysis, oil analysis, thermography, ultrasound, motor circuit testing and the rest, whoever performs them, operators and contractors included. Condition-based, on-condition and predictive maintenance all describe the same idea here: you act on the evidence of the machine's actual health rather than the date on a calendar.

One distinction does the metric's heavy lifting, and it is the one people most often get wrong. The cost covers the monitoring itself. The repair that a finding triggers is corrective work, and it belongs in the corrective column, not here.

How to work it out

Condition-based maintenance cost = (CBM cost ($) × 100) / Total maintenance cost ($)

In a month, predictive work orders total $17,100, contractor predictive work $9,300 and operator monitoring $4,898, against a total maintenance cost of $194,400.

CBM cost = ($31,298 × 100) / $194,400 = 16.1%

About sixteen cents in every maintenance dollar went on listening rather than fixing. On its own that figure means little; tracked beside the corrective cost share, quarter after quarter, it starts to talk. A falling corrective line sitting behind a steady or rising CBM line is the signal that the investment is doing its job.

The window that makes it pay

The whole idea rests on a window engineers call the P-F interval. P is the point at which a failure first becomes detectable, the earliest whisper a sensor can pick up; F is the point at which the equipment actually stops. Everything condition monitoring buys you lives in the gap between the two. Catch the warning early in that window and you convert a sudden breakdown into a repair you can plan, schedule, resource and carry out at a fraction of the cost and disruption. Miss it and you are back to reacting.

That window is also why how often you monitor matters as much as whether you monitor at all. The rule of thumb is to take readings at no more than half the P-F interval, so no developing fault can run its full course between two checks. A bearing that gives ninety days of warning needs looking at least every forty-five; sample it quarterly and you may walk straight past the very failure you were paying to catch.

The honest case for the spend

It is tempting to sell condition monitoring on its headline successes, and the headline successes are real: the right technique, on the right asset, at the right interval, can return anywhere from five times its cost to more than a hundredfold. But the honest picture is more sobering, and worth knowing before you build a business case on the dream figures. Across all the plants that run predictive programmes, the average return is barely better than breaking even, and a large share never recover what they spend. Only a minority, perhaps one programme in ten, earns the spectacular returns the technology is capable of.

The gap between those two stories is not the technology. It is execution. The dramatic returns go to the plants that monitor the right equipment, at the right interval, and then actually act on what they find, closing the loop with a planned repair before the failure lands. The break-even programmes buy the sensors and the software and then let the findings pile up unactioned, or spend their effort monitoring low-consequence assets that would have been cheaper to run to failure. CBM cost tells you what you are spending. Whether that spend joins the top tenth or the disappointing average depends entirely on the discipline behind it.

What good looks like

There is no formal benchmark for the cost version, and it moves closely with CBM hours, where a common working guide puts condition-based work somewhere around a sixth of total maintenance effort. Read the two together. If cost climbs faster than hours, dig into whether contractor rates or a new piece of monitoring technology is driving it, rather than a genuine increase in the work itself. One further cross-check is worth building: the yield of the monitoring, roughly how many hours of justified corrective work each hour of inspection turns up. A healthy programme tends to generate somewhere between half an hour and two hours of warranted repair for every hour spent looking. Far below that and you may be monitoring assets that do not need it; far above and your inspections may be running too late to catch things early.

Where it can mislead you

• CBM cost is the monitoring only. The repair that follows a finding belongs in corrective cost, and quietly relabelling those repairs as CBM will flatter this figure while hiding the real reactive burden, which is exactly the wrong way round.
• Keep a consistent time basis. A turnaround can spike total maintenance cost in a given month and temporarily squash the CBM percentage, which says nothing at all about your actual monitoring effort.
• Your work order system has to separate CBM from PM and corrective work cleanly, or the number cannot be trusted.
• Include operator monitoring costs if operator maintenance already sits in your total, so the top and bottom of the fraction stay consistent.
• Beware the empty programme. A site that claims to run predictive maintenance while booking almost nothing to CBM cost is not really running one, and the missing spend is a finding in itself.

CBM cost is the price of early warning, and it is one of the few line items on the maintenance ledger whose whole purpose is to make another one shrink. Watched alongside your corrective spend, and backed by the discipline to act on what the monitoring finds, it turns the nagging question, are we getting anything for all this listening, into one the trend can finally answer.