Corrective maintenance: how it's managed

Corrective maintenance is what you do once something has already failed. It has a bad reputation, and doesn’t always deserve it: for cheap, redundant equipment, or equipment whose failure has no consequence, it’s the right strategy. Preventively checking a lightbulb costs more than replacing it when it burns out.

The problem isn’t having corrective maintenance. It’s when it’s all there is, and when it happens without being recorded.

The two types, and they’re not managed the same way

Scheduled corrective. The failure has been detected but doesn’t force a stop: an anomaly spotted during a check, a component that’s working worse than it should. It gets planned, the material is ordered, and it’s done when it’s convenient.

Urgent corrective. The equipment is down and someone has to go now. This is the expensive kind: immediate travel, a part with no lead time, rushed work, and often, other work that goes undone.

The ratio between the two says a lot about an operation. If almost everything is urgent, the preventive plan isn’t catching anything.

The workflow, step by step

1. The alert comes in through one place. An incident with its type and subtype, priority, client, site, and affected equipment. It can arrive from the backend, from client access with a description and photo, or from a mailbox that the system converts into alerts.

As long as there are three parallel channels, there’s no reliable pending list.

2. It gets prioritized by criteria, not by who shouts loudest. Statuses can carry their own maximum time, and there’s an SLA entity with its priority and its limit. That way overdue items show up in a list instead of being discovered when the client complains.

3. It turns into an order. From the incident comes the work order, carrying over the client, site, and equipment, with its technician and date. The order keeps its origin, so you can always trace back from the job to the alert that triggered it.

4. The technician arrives with context. In the app they have the equipment’s history, its documentation, and any anomalies left open. Many breakdowns are repeats of something already seen, and knowing that before starting is the difference between fixing it and patching it again.

5. What happened gets recorded. Time tracked with a timer inside the order, material consumed against the warehouse, photos, cause of stoppage, and signature. It all works without signal and syncs afterward.

6. It gets closed, or explicitly left open. The order can be flagged as pending material and as needs a return visit. Without those states, that work lives in a notebook, and the client remembers before we do.

The data you have to capture, no exceptions

If all that’s left of a breakdown is “it got fixed,” it’s useless afterward. The minimum is this:

The cause of the stoppage, which is its own field on the order and what later lets you analyze why something keeps happening.

The actual time, measured. The order stores the target time, the estimate, and the deviation, which reveals which type of job is being budgeted wrong.

The material, with its lot number when it has one.

Whether it was under warranty. The asset record carries its warranty end date, and the order can be flagged as covered work. Paying for repairs on equipment still under warranty is a silent, avoidable cost.

How to reduce urgent corrective work

Not by checking more often, which is the intuitive and wrong answer. It’s reduced through three routes.

Detecting earlier. A checklist with a minimum and maximum value turns a check into a measurement: a reading out of range gets logged as an anomaly on the spot, and that anomaly turns into scheduled corrective work instead of an urgent breakdown three months later.

Adjusting periodicity with history. If a piece of equipment breaks down between checks, the frequency is set wrong. And if it’s gone several cycles without a single anomaly, a visit is probably unnecessary. Both cost money, and the only way to tell them apart is with data.

Changing the trigger when it’s warranted. If wear depends on use rather than time, the asset can carry a counter with a limit and a warning percentage, so the preventive order generates automatically once the threshold is crossed.

When to stop repairing

The most expensive question in corrective maintenance, and the one almost always answered by gut feeling.

With accumulated cost per asset in front of you — hours plus material, over years — the conversation changes. The reports on cost per equipment, downtime, MTBF, and MTTR are what back it up.

And there’s an automation built for exactly that: an asset can have its replacement cost and a warning percentage recorded, and the system notifies when accumulated repair spend exceeds it. It doesn’t generate an order — replacing a piece of equipment is a business decision — but it puts the figure in front of you right when it matters.

The indicators

MTTR, mean time to repair: measures response capability.

MTBF, mean time between failures: measures reliability. These are two different things, and something can be repaired very fast while still failing too often.

Accumulated downtime per piece of equipment, which translates the above into lost production or service.

Ratio of corrective to preventive hours, which best describes whether the operation is running behind. It takes months to move, which is exactly why it’s honest.

The material, which is half the urgency

A large part of what turns a breakdown into a problem isn’t the repair: it’s that the part is missing. The visit stalls halfway, someone has to go back, and the equipment stays down while the part arrives.

Three things reduce this. Minimum stock per item in warehouses and items, which warns before you run out of what a job needs. Registering each technician’s van as a warehouse, because that’s where the material that actually travels lives. And the item’s original reference, which lets you find another manufacturer’s equivalent when the original has weeks of lead time.

The list of what’s worth keeping in stock comes from breakdown history after a few months, not from an initial estimate.

What not to do

Eliminate corrective maintenance from the plan. You can’t, and you shouldn’t: there are sudden failures that give no warning, and equipment where repairing is cheaper than preventing.

Manage it outside the system. If emergencies get resolved by phone and never get recorded, the history has gaps exactly where the most expensive interventions are.

Measure the technician by MTTR. It’s a process indicator, not a personal one. As soon as it’s used to evaluate people, the recorded times stop being real.

There’s more product detail on the corrective maintenance page. If you want to see it applied to your operation, you can request a demo.