Corrective Maintenance: When Fixing It After It Breaks Is the Right Call

Corrective maintenance is repair work done after a fault is found, to bring equipment back to its working condition. It covers everything from a planned fix of a worn seal spotted on an inspection to an emergency callout for a pump that stopped mid-job. Most maintenance guides treat it as the thing to eliminate. For a good share of the equipment in a service or rental fleet, it is the cheapest sensible plan, provided the work gets recorded.

This guide explains what corrective maintenance is and is not, the types you will see in standards and in the field, when running a unit to failure is a decision rather than a failure, how to price it against a preventive visit using your own numbers, and how to pick the assets where it belongs. For where it sits among the other strategies, see our comparison of preventive, predictive and proactive maintenance.

Key takeaways

• Corrective maintenance is a strategy when you choose it for cheap, non-critical, obvious-failure assets, and a cost when it chooses you.

• Price it per unit and per site: a quarterly check that rides along on a visit already happening costs a fraction of one that needs its own trip.

• Record every corrective job with eight fields, including the meter reading, or you will never know when a run-to-failure asset has stopped being cheap.

What Corrective Maintenance Is (and Is Not)

The terms overlap, and they are used inconsistently: the U.S. Department of Energy's O&M Best Practices guidance treats reactive and corrective maintenance as the same thing, while many software vendors treat reactive as the unplanned subset of corrective. This is how the words are used in this guide.

The useful distinction is between corrective work you chose and corrective work that chose you. The first is a strategy. The second is a cost.

Corrective vs preventive maintenance

Most operations run both and choose by asset. The preventive, predictive and proactive maintenance guide covers the scheduled side in depth.

Types of Corrective Maintenance

The European maintenance terminology standard, EN 13306, splits corrective maintenance by timing:

• Immediate corrective maintenance: carried out without delay after a fault is detected, to avoid unacceptable consequences. The pump that stopped mid-job.

• Deferred corrective maintenance: the fault is detected, but the repair is delayed to a planned time. The weeping seal noted on Monday and replaced when the unit is back in the yard on Friday.

Planned corrective maintenance covers both run-to-failure repairs and faults found during inspections or preventive visits. In the field, the common version of deferred work is a swap: the failed unit comes off the job, a ready unit goes on and the repair happens in the shop on a repair order rather than on location.

Maintenance handbooks also split corrective work into five kinds, each with an obvious field equivalent:

When Run-to-Failure Is a Strategy, Not a Failure

The DOE guide is careful here. It calls the advantages of reactive maintenance a double-edged sword, and in the same breath notes that work on inexpensive equipment that does not matter to facility reliability may be best left reactive. Its Table 5.5.1, adapted from NASA, lists where that applies: small parts and equipment, non-critical equipment, equipment unlikely to fail and redundant systems. The same conclusion comes out of a reliability centered maintenance analysis: for some failure modes, the right task after analysis is no task at all.

An asset is a candidate for run-to-failure when most of these hold:

• Failure does not create a safety or environmental exposure.

• Failure does not stop a job, a revenue asset or another crew's work.

• Failure is obvious to the operator the moment it happens. Protective devices with hidden failures, such as H2S monitors, high-level alarms, pressure switches and standby pumps, are excluded: nobody knows they have failed until they are needed. These get a scheduled function test, which is an RCM task, not run-to-failure.

• A spare or a replacement unit is on hand or available the same day. Our guide to oil and gas inventory management covers how to keep that true on trucks and in yards.

• The repair or replacement costs less than a year of preventive visits (the break-even math below).

• Failure gives little or no warning, so inspections would not catch it anyway.

• A redundant unit covers the function while this one is down.

• The unit is near the end of its life or scheduled for disposal, so preventive spend will not be recovered.

Two exclusions override the cost math every time. First, compliance: equipment subject to DOT annual inspections, OSHA forklift and crane inspection rules, API or OEM recertification or a warranty that requires documented preventive maintenance is not a run-to-failure candidate whatever the numbers say. Second, anything on the critical path of a job or a safety system.

Shop drain pumps, work lights, hand tools, basic sensors with no protective role, consumable filters and most office equipment usually qualify. A pressure pumping unit, a crane or a generator powering a site do not, and never will.

When Corrective Maintenance Is Cheaper: The Break-Even Math

Program-level statistics do not settle this for a single unit. The DOE guide reports that studies from around 2000 found more than 55 percent of maintenance resources at an average facility going to reactive work, and that consistently top-performing facilities run under 10 percent reactive. Note the second figure: even the best programs keep some corrective work by design. The question is which units.

Price each option per year using the unit's own numbers. Cost per failure is parts, plus labor, plus travel or callout, plus downtime or lost revenue, plus any secondary damage. Here are illustrative numbers for a $400 shop drain pump with $100 an hour loaded labor and a failure about every two years:

The third row is the one to notice. The same pump flips from run-to-failure to preventive maintenance depending on whether a crew is already on site, so the decision is made per site, not per asset type. A drain pump in the yard where a technician walks past it daily gets a check; the same pump at a remote lease does not.

Now run the same math on a revenue asset. Put a day of lost rental or job revenue into the cost per failure for a site generator or a frac pump and run-to-failure fails the test immediately, before you even count secondary damage. That is the whole method: cheap units with no consequence run to failure, and anything that stops work does not.

Corrective Maintenance Examples: A Decision Table for Field Equipment

Score each unit on two things: what happens when it fails, and how much warning it gives. The result points at a starting strategy. The table below covers ten common units in oilfield and rental fleets; the method works for any asset list.

A note on the consumables row. Standards put replacing a worn but still working belt under condition based maintenance (and the DOE table lists consumable and wearing equipment under preventive), while many teams log it as planned corrective. Pick one and apply it consistently, because the choice changes your PM-to-corrective ratio and every KPI built on it.

The bottom four rows are where corrective maintenance earns its place. Review the list once a year; units move up the table as the fleet and the contracts change.

Not sure which of your units are on the wrong row? Run the 2-minute operations assessment and see where the leaks are. Take the assessment

Pros and Cons of Corrective Maintenance

How to Run Corrective Work Without Losing the Record

The failure mode of a run-to-failure policy is not the failure. It is the repair that never gets written down. A corrective job, planned or not, should follow the same eight steps and leave the same record:

1. Report: the operator or the inspection logs the fault against the unit, with a photo.

2. Triage: is it immediate or deferred? Can the job continue with a workaround or a swap unit? Breakdown callouts run through dispatch like any other job.

3. Dispatch or swap: send a technician, or send a replacement unit and bring the failed one to the shop on a repair order.

4. Diagnose: confirm the failure mode before ordering parts.

5. Repair, test and return to service, with the return-to-service note on the record.

6. Close with codes: what failed, what was found, what was done.

7. Bill: mark whether the cause is customer damage or normal wear, so the rental or service invoice is right.

8. Review: once a quarter, count corrective work per unit and per part.

The record that makes step 8 possible has eight fields. Digital work orders capture them in the field; a paper ticket almost never does.

1. The asset, by its unique ID, not by description.

2. What failed and what was found, from a short failure code list. Oil and gas teams can seed the list from the failure mode taxonomy in ISO 14224.

3. Parts used, drawn from inventory so usage history builds itself.

4. Labor hours, including travel if the unit was in the field.

5. Downtime: when it stopped and when it was back in service.

6. Meter reading at failure: hours, miles or cycles. Without it you cannot see parts failing at the same run hours across the fleet.

7. Whether the failure stopped a job or caused a callout.

8. Billable to the customer, as damage or normal wear, or absorbed.

If a run-to-failure drain pump has failed three times this year and each time cost a four-hour callout, it has stopped being a cheap asset. A system that links the work order to the asset and the parts is what makes the count possible; our guide to CMMS, EAM and FSM explains which system does which job.

Corrective Maintenance for Rental Fleets

Rental changes the math in two ways. A down unit is off rent, so the cost per failure includes the daily rate for every day it sits, not just the repair. And the cause decides who pays: damage is billable to the customer, normal wear is not, which is why the eighth field above exists. Rental teams also carry more units that fit run-to-failure (hand tools, small pumps, lights) and more that never will (anything with an inspection certificate). Our equipment rental business software buying guide covers what the rental record has to hold for this to work.

Corrective Maintenance KPIs

Five numbers tell you whether the strategy is working. Two of them, mean time between failures and mean time to repair, have their own formulas and worked example.

• MTBF per unit: is the run-to-failure asset failing more often than you assumed?

• MTTR and mean downtime: how long a corrective job takes, and how long the unit is actually out.

• Corrective cost per asset per year: parts, labor, travel, downtime and lost revenue, from the work orders.

• Repeat-failure rate: the share of units with more than one corrective work order for the same failure code in 12 months.

• PM-to-corrective ratio: how many preventive visits you run for each corrective job found.

On that last ratio, a common rule of thumb is 6:1. Preventive inspections should surface a corrective job about once every six visits. Much above 6:1 and you are inspecting too often (five PM visits and no findings is wasted money); much below it and not often enough. Use three triggers to move an asset back from run-to-failure to a preventive schedule: three failures in 12 months, any failure that stopped a job or an annual corrective cost above the cost of a PM program for that unit.

Where AI Fits

AI does not change the strategy for a $400 pump. It changes how fast you notice the strategy is wrong. E-Genie, Equipt's AI assistant, lets you ask questions in plain language about equipment status, performance and field activity, such as: "Which run-to-failure assets logged more than two corrective work orders in the last 12 months?" or "Which part failed on more than one unit at similar run hours this year?" The reverse question is just as useful: which units had five preventive visits and no findings in two years? That is the sign the preventive money is wasted.

That is the practical version of corrective maintenance done well: cheap by design, recorded every time and reviewed by someone or something that counts. Equipt's field service management software tracks every work order, corrective or preventive, from creation to closure with time-stamped evidence, so the review happens on data rather than memory.

Frequently Asked Questions

What is corrective maintenance?

Corrective maintenance is repair work carried out after a fault has been identified, to restore equipment to working condition. It can be planned (a worn part found on inspection is replaced on the next visit) or unplanned (the equipment has already failed).

What are the types of corrective maintenance?

By timing: immediate (repaired without delay) and deferred (repair planned for later), as defined in EN 13306. By the kind of work: fail-repair, servicing, overhaul, rebuild and salvage. Planned corrective maintenance covers run-to-failure repairs and faults found during inspections.

What is an example of corrective maintenance?

Replacing a dead starter on a light tower after it fails to start is unplanned corrective maintenance. Replacing a weeping hydraulic hose found on a pre-trip check, at the next yard visit, is planned (deferred) corrective maintenance.

What is the difference between corrective and preventive maintenance?

Preventive maintenance is done on a schedule before a fault appears, to reduce the chance of failure. Corrective maintenance is done after a fault appears, to fix it. Most operations use both, choosing by how critical each asset is.

Is run to failure the same as corrective maintenance?

Run to failure is a policy: no preventive work on an asset, and corrective maintenance when it fails. So run to failure always leads to corrective maintenance, but corrective maintenance also happens on assets that do get preventive work.

How do you decide which assets to run to failure?

Pick units whose failure causes no safety or environmental exposure, does not stop a job, is obvious when it happens and costs less to fix than a year of preventive visits, with a spare on hand. Exclude anything with an inspection, certification or warranty requirement. Then price it per site, because a check that rides along on a visit already happening is cheap.

What is breakdown maintenance?

Breakdown maintenance is unplanned corrective maintenance performed after equipment has stopped working. It is the most expensive form of corrective work because it usually involves downtime, callouts and expedited parts.

Does the 80/20 maintenance rule apply to every asset?

No. An 80 percent preventive to 20 percent corrective split is a program-level rule of thumb. A fleet with many small tools and non-critical units can legitimately run more corrective work; a fleet of revenue-critical equipment should run less. Set the ratio per asset class, not for the whole operation.

Want to see which of your units are quietly costing more in corrective work than a PM would? Book a demo