Key takeaways
Short answer: Time-based maintenance services an asset on a fixed calendar or runtime interval whether it needs it or not. Condition-based maintenance acts only on evidence, vibration, temperature, oil analysis, that the asset is degrading. Time-based is simple but over- or under-services; condition-based targets real need but requires sensing and analysis. The right mix is decided by failure pattern, criticality and monitoring cost. See also preventive vs corrective maintenance.
Time-based maintenance runs to a schedule: every 500 hours, every quarter, every so many cycles. It is simple to plan, easy to audit, and works well where failure follows a predictable age pattern. Its weakness is that the interval is a guess, too short and you over-service, too long and you miss failures.
Condition-based maintenance (CBM) watches the asset and intervenes only when measured condition says it is degrading. Instead of guessing the interval, it reads the evidence, and does the right work at the right time, provided you can sense the right signal.
A gearbox is serviced every six months on a time-based plan. Two of those services replace healthy oil and seals, pure waste, while a bearing that started failing in month four runs to destruction in month five, before the scheduled service.
Add a vibration sensor and the same gearbox is left alone while healthy and flagged the moment the bearing signature changes. CBM removed two wasted services and caught the one failure the calendar missed.
For wear-out failures with a clear age pattern, time-based is fine and cheap. For random or hard-to-predict failures on critical assets, condition-based earns its monitoring cost. Most plants run a mix decided by criticality, not a single doctrine.
1. Time-based everything. You waste labour over-servicing stable assets and still miss random failures.
2. CBM without a clear signal. Monitoring a failure mode you cannot actually detect adds cost and false confidence.
3. Ignoring criticality. Spending CBM money on trivial assets and running critical ones on a calendar.
4. Collecting condition data nobody acts on. A sensor whose alerts are ignored is just expense.
Both strategies aim to protect Availability. Condition-based, done well, cuts both unplanned breakdowns and unnecessary PM downtime, lifting OEE from two directions, fewer surprises and less time on the calendar servicing healthy machines.
Fabrico ties condition signals and downtime together so you can see which assets justify condition-based care and prove the Availability gain. Book a demo to see condition and downtime in one place.
No, for predictable wear, time-based is simpler and cheaper. Match the strategy to the failure pattern.
A sensible signal to sense, plus the analysis and discipline to act on it.
Yes, most plants assign strategy by asset criticality rather than picking one doctrine.
Predictive maintenance is CBM plus failure-time forecasting; CBM is the foundation it builds on.
On critical assets with random failure modes and a clear, detectable signal.
Login
Blog
