Corrective maintenance is any maintenance work done to restore a failed or failing asset to working condition. It splits into two very different kinds: planned corrective work, scheduled after a problem is found but before it stops the line, and emergency corrective work, done in a hurry after something has already broken. Telling them apart is the whole point of managing it.
Corrective maintenance has a reputation problem. People hear it and picture a technician sprinting to a dead machine at 3 a.m. That is one kind of corrective maintenance, the worst kind, and treating it as the whole category hides the fact that most corrective work can and should be planned. This guide defines the term properly, splits planned from emergency, shows where each comes from, and lays out how to keep the balance on the right side.
What is corrective maintenance?
Corrective maintenance is maintenance carried out to identify, isolate, and fix a fault so the asset can return to a condition where it can perform its function. It is the counterpart to preventive maintenance: preventive work happens before a fault exists to stop it forming, and corrective work happens because a fault already exists or is developing.
The key word is fault not failure. A fault can be a bearing that is running hot, a seal that is weeping, a guard that is cracked, or a pump losing efficiency, none of which has stopped production yet. Corrective maintenance covers fixing all of those, plus the outright breakdowns. That range is exactly why the category needs splitting: fixing a weeping seal on Saturday during planned downtime and replacing a seized pump mid-shift are both corrective maintenance, but they are not the same event in any way that matters to a budget or a schedule.
Is corrective maintenance the same as reactive maintenance?
No, reactive maintenance is a subset of corrective maintenance, specifically the emergency, run-to-failure part. All reactive maintenance is corrective, but not all corrective maintenance is reactive. This is where a lot of confusion lives, so it is worth drawing clearly.
The reason this matters is that the two branches behave like different jobs. Planned corrective work is scheduled into a window, parts are staged, the right technician is assigned, and lockout is prepared. Emergency work interrupts whatever the crew was doing, often runs without the right parts on hand, and gets done under time pressure, which is exactly when mistakes and injuries happen. When someone says a plant is too reactive, they mean too much of its corrective work is landing on the emergency branch.
What is the difference between planned and emergency corrective maintenance?
Planned corrective work is fixed on your schedule; emergency corrective work is fixed on the machine's. Everything else follows from that one difference.
| Dimension | Planned corrective | Emergency corrective |
|---|---|---|
| Trigger | Fault found during inspection, PM, or condition monitoring | Sudden failure that stops the asset |
| Timing | Scheduled into a maintenance window | Now, whenever it happens |
| Parts | Staged in advance | Whatever is on the shelf, or a rush order |
| Labor | Right technician, planned hours | Whoever is available, often overtime |
| Safety | Proper lockout, no time pressure | Time pressure raises injury risk |
| Production impact | Minimal or none | Line down, often on the worst day |
| Cost | Baseline | Often several times higher |
A worked example makes it concrete. Say vibration analysis flags a failing gearbox bearing with weeks of warning. As planned corrective work, you order the bearing, wait for a scheduled window, lock out the drive, and swap it in ninety minutes with the line already down for changeover, near-zero production impact. Miss that warning and the bearing seizes mid-shift: now it is emergency corrective work, the line is down, you are paying overtime and possibly expediting a part, the repair is rushed, and you may have taken out the shaft on the way. Same bearing, wildly different event.
Where does corrective maintenance come from?
Planned corrective work is generated by everything you do to look at equipment before it fails; emergency work is generated by the gaps in that looking. The best corrective-maintenance programs are downstream of good detection.
Planned corrective work is fed by three sources. Preventive maintenance findings a technician doing a scheduled PM notices a worn wear strip or a leaking fitting and writes it up as corrective work. Inspections and operator checks first-line care under total productive maintenance turns operators into the eyes that catch developing faults early. Condition and predictive monitoring vibration, oil analysis, and thermography under predictive maintenance flag degradation with lead time to plan the fix. Every one of these converts a would-be breakdown into scheduled corrective work.
Emergency work comes from what those methods miss: random failures, faults that develop faster than the inspection interval, and the assets you deliberately chose not to monitor. Some emergency work is unavoidable, no program catches everything, and for cheap, redundant assets, running to failure is the right economic choice. The problem is not that emergency work exists; it is when it dominates, because that means detection is failing or PM is not being done.
How does corrective maintenance fit the overall strategy?
Corrective maintenance is not a strategy you choose against preventive or predictive maintenance, it is the work that happens when a fault exists, regardless of how you found it. Every maintenance strategy produces corrective work; the strategy just decides how much of it is planned versus emergency.
Think of it on the reliability ladder in our equipment reliability guide. A purely reactive plant does almost all corrective work as emergencies. Add a real preventive maintenance schedule and PMs start catching faults, so corrective work shifts toward planned. Add condition-based and predictive monitoring and even more shifts left, with more warning. Run-to-failure still has a place for low-consequence assets, a deliberate, documented decision, not a default. The maturity of a maintenance organization is visible in one ratio: what fraction of its corrective work is planned.
How do you manage corrective maintenance well?
Manage it as a workflow with the same discipline as planned work: capture the fault, prioritize it, plan it, schedule it, execute it, and close it with a cause. The sequence is what keeps found problems from becoming forgotten problems.
- Capture every fault as a work request. The moment anyone, technician, operator, quality, sees a developing problem, it becomes a written request with the asset, the symptom, and how urgent it looks. Faults kept in someone's head are the raw material of tomorrow's breakdown.
- Triage: emergency or planned? Decide immediately whether it must be fixed now or can wait for a window. Use a simple rule based on safety, production impact, and how fast the fault is progressing. Most faults, honestly assessed, can wait, which is the whole opportunity.
- Prioritize the planned corrective queue. Rank by asset criticality and consequence, not by who shouted loudest. This queue is your maintenance backlog and a healthy backlog is a sign of good detection, not bad execution, provided it is sized in crew-weeks and actually worked down.
- Plan the job. Identify parts, tools, skills, permits, and duration before it is scheduled. Planning is what separates a two-hour window that goes smoothly from a four-hour one waiting on a part, the core of good maintenance planning and scheduling.
- Schedule it into a window. Slot planned corrective work into maintenance windows, changeovers, or planned downtime so it costs the least production. Protect those windows the way production protects a run.
- Execute and record what was actually done. Capture parts used, time taken, and findings against the asset history. This record is what makes the next similar fault faster to fix and feeds your mean time to repair.
- Close the loop on repeat failures. When the same fault keeps coming back, run a root cause analysis instead of replacing the same part forever. Corrective maintenance that never asks why is just organized firefighting.
What does corrective maintenance cost, and what should you measure?
The cost gap between planned and emergency corrective work is the entire argument for shifting the balance, and the public maintenance economics back it up.
- The U.S. Department of Energy's Federal Energy Management Program guidance, maintained by PNNL, documents that typical facilities spend a large share of maintenance effort reactively, while best-in-class operations hold reactive work to a small fraction, and that moving from a reactive posture toward planned maintenance recovers a substantial share of maintenance cost (PNNL, O&M Best Practices: Maintenance Approaches).
- The same guidance estimates a functioning preventive maintenance program saves meaningfully over a reactive one, and predictive techniques save further still, savings that come precisely from converting emergency corrective work into planned corrective work (DOE FEMP, O&M Best Practices Guide).
- Emergency corrective work also consumes your scarcest resource, skilled technicians. The U.S. Bureau of Labor Statistics projects employment of industrial machinery mechanics and millwrights to grow about 13% from 2024 to 2034 much faster than average, with roughly 54,200 openings a year (BLS Occupational Outlook Handbook). Every hour spent firefighting is an hour not spent preventing the next fire.
The metrics that tell you whether corrective maintenance is under control: the percentage of maintenance work that is reactive (lower is better; best-in-class runs well under 20%), mean time to repair the ratio of planned to emergency corrective work and backlog in crew-weeks. The full set is in our guide to maintenance KPIs. And because emergency corrective work is where machine downtime and its full cost come from, those numbers are the bridge between the maintenance office and the plant P&L.
Where does software fit?
Corrective maintenance falls apart in the gaps: a fault someone noticed but never wrote down, a work request that sat unprioritized, a repair whose history was never recorded. Those gaps are information problems before they are maintenance problems. A system that captures every fault the moment it is seen, triages and schedules it, and holds the asset history is what keeps corrective work from decaying into firefighting.
This is the shift Harmony supports on the floor: operators and technicians log faults and complete work on tablets at the asset, machine signals surface developing problems, and the history is searchable instead of stacked on paper. It layers onto the ERP, MES, and machines already in place. No rip-and-replace. The CLS case study shows a plant moving from paper logs reviewed the next morning to catching and acting on issues during the same shift, and the platform overview shows how it connects. The target is simple: make planned corrective the default, and keep emergencies rare.