Maintenance: Fixed Intervals or Condition Evidence Before Sensors?
Choosing time-based or condition-based maintenance starts with failure mechanisms, detectability and consequences. Buying a sensor is not the same as building a reliability program, and a schedule does not guarantee prevention.
The Idea in One Sentence
The real choice is not whether to replace calendars with sensors. It is how to detect deterioration early enough at acceptable cost and residual risk.
Why It Matters
Factories contain failure modes with different behaviors. A seal may deteriorate gradually; an electronic component may fail without useful warning. Distinguish observable degradation from unpredictable breakdowns and set safety obligations as nonnegotiable.
Explain It Simply
A pump is overhauled every six months despite low operating hours. Should the interval change or should vibration be monitored? Without failure history, reliable measurements and response capacity, both options are largely guesses.
Evidence Map
The US Department of Energy distinguishes reactive/corrective maintenance, time-based prevention, condition-related prediction and reliability-centered selection of appropriate tasks. Its guidance does not establish universal savings for every asset or excuse mandatory inspections.
Core Mechanics
Define the asset, function and failure modes. For each failure, examine consequence, warning signal, detection-to-failure interval, signal quality and ability to act. Then compare timed tasks, condition monitoring, protective controls and other management choices. Safety functions may require periodic testing regardless of sensor data.
Flows and Responsibilities
Operational flow goes from machine to measurement, diagnosis, work order, repair and confirmation. Information includes failure history, false alarms, downtime and post-repair outcomes. Financial flow covers planned labor, sensors, avoided disruption, repairs and lost production. Measurements need to support action.
Economic Logic
Compare future scenario costs: repeated inspections, consumables, monitoring, calibration, analysis, avoided interventions and missed failures. Fewer scheduled repairs do not prove economic improvement if one catastrophic failure remains possible.
Constraints and Failure Modes
Noisy signals, missed detections, unavailable technicians, long-lead parts, statutory inspection requirements and poor configuration can defeat the business case. Data capture without response process adds a dashboard to failure instead of preventing it.
What Most People Miss
Technology is not the first decision. The relevant capability is detection → diagnosis → resource availability → verified action. A cheap sensor cannot fix an unobservable failure mode, while simple visual inspection can be superior for some risks.
Critical View
Performance depends on equipment duty and failure modes. DOE guidance provides maintenance principles, not observed savings at a particular site. Avoid universal savings claims without comparable equipment, time horizon and denominators.
Sidy’s Synthesis
My synthesis: a maintenance strategy buys actionable warning time before consequences occur. Ask four questions: is deterioration observable, is the warning early enough, can someone act in time and what risk remains? The right instrument is the one that changes a real decision, not the one generating the most data.
AI & Future Lens
Today AI can classify historical signals but may infer spurious causes. Multisignal detection could improve in five years. In ten years maintenance and parts sourcing may be better aligned. In twenty years new equipment will have new failure modes. None removes the need for field validation.
Build From This
Build a three-asset maintenance choice sheet without buying equipment. Fields: failure, consequence, observable sign, warning lead time, available action, current cost and proposed cost. Replay six past incidents and reject monitoring if it would not have changed an intervention.
Practical Actions
Choose one asset. Review its three latest failures. Verify actual parts lead time. Compare a planned task with a condition signal and record the decision rationale.
Remember This
Sensor ≠ actionable prediction. Periodic maintenance ≠ guaranteed prevention. The test is whether action can precede harm.
Primary sources
Facts, figures and quotations should be traceable to the sources below. Sidy's synthesis is labeled as synthesis and does not replace sourced facts.