There is no universal preventive-maintenance interval that is correct because an asset is a pump, motor, conveyor, fan, or gearbox. The interval should follow the failure mode, task type, deterioration behavior, detection method, governing authority, consequence, operating exposure, and evidence from actual maintenance history.
The task may be incapable of detecting the failure, the failure may be random, the inspection method may be weak, the repair may not address the contributor, or the better strategy may be condition monitoring, redesign, failure-finding, or run-to-failure.
When someone asks me whether a monthly PM should become weekly, my first question is not about the calendar. It is: what failure are we trying to control, and what evidence tells us this task can see or prevent it? If that answer is weak, changing the interval is usually premature.
Regulatory, safety, quality, insurer, approved engineering, or OEM requirements may establish a hard constraint unless formally revised through the appropriate authority.
If failure probability changes meaningfully with age or usage, a time- or usage-based task may be defensible when the interval is supported by evidence.
If deterioration can be detected with enough warning to act, condition-based maintenance may be more effective than a fixed calendar interval.
Does the task routinely find actionable deterioration, or does it repeatedly close without useful condition information?
Do relevant failures occur before the next PM, immediately after completed PM work, or without a meaningful age pattern?
When PM identifies a defect, does corrective work occur before functional failure? A good inspection interval is useless if findings are not acted on.
Does performing the PM itself create risk through disassembly, contamination exposure, adjustment error, or unnecessary maintenance-induced failure?
Frequency changes affect labor demand, equipment access, production interruptions, permits, sanitation, and coordination. More frequent is not free.
Higher-consequence failures justify more rigorous evidence, detection methods, redundancy, or engineered controls than low-consequence failures.
If a failure mode produces a detectable potential-failure condition before functional failure, the inspection interval should consider the time between detection and loss of function, the variability of that interval, the reliability of the detection method, consequence, access, and the time needed to plan and execute corrective work.
Can the method reliably identify the deterioration at the stage where intervention remains practical?
After a finding, can the organization obtain parts, access, permits, labor, engineering, and downtime before failure?
Higher consequence or highly variable deterioration may justify a more conservative interval, redundancy, continuous monitoring, or another control.
| Finding | Possible disposition |
|---|---|
| Task detects deterioration too late | Shorten interval, improve method, or move to a more sensitive condition-monitoring technique. |
| Task repeatedly finds no meaningful condition and risk is low | Review whether interval can be extended or task removed, subject to technical authority. |
| Failure remains random and not detectably degrading | Consider redesign, spares, operating control, or run-to-failure rather than more calendar PM. |
| PM finds defects but failures still occur | Investigate follow-up execution, acceptance criteria, repair quality, and whether the task covers the correct failure mode. |
| Task is mandatory | Maintain the required interval unless the governing authority formally changes it. |
| Evidence pattern | Question | Possible action |
|---|---|---|
| Defects are routinely found too late | Is the method sensitive enough, and is the inspection interval shorter than the useful warning window? | Improve method, shorten interval, add online monitoring, or reduce planning response time. |
| Task almost never finds deterioration | Is the failure mode still credible? Is the task capable? Is risk low enough to extend or remove? | Extend, revise, convert strategy, or remove with technical review. |
| Failures occur immediately after PM | Could intrusive maintenance, workmanship, setup, contamination, or restoration quality be contributing? | Improve precision standard, reduce intrusiveness, revise method, or investigate maintenance-induced failure. |
| PM finds defects but failures still occur | Are findings converted into corrective work and completed before the functional failure? | Fix follow-up process before changing inspection frequency. |
| Failure timing is random and no detectable precursor exists | Does calendar PM have technical leverage over the mechanism? | Consider redesign, spares, operating controls, or deliberate run-to-failure. |
| Interval is mandatory | Who has authority to change it? | Maintain requirement or pursue formal approved revision. |
For each task, record the failure mode, task purpose, current interval, interval basis, source or authority, relevant history, proposed disposition, confidence, and required review. This prevents future teams from inheriting a number with no explanation.
A PM frequency without a documented basis is just an inherited number. The goal is not to make intervals longer or shorter. The goal is to make them defensible.
Evaluate PM task quality, duplication, technical basis, and failure-mode coverage.
Choose PM, PdM, failure-finding, redesign, or run-to-failure based on the failure mechanism.
Separate schedule execution from whether PM work actually controls failure.
Use the method yourself, or apply the same reasoning to your facility's CMMS history through XRVE Reliability.
Free maintenance maturity assessment See sample analysis