Predictive Maintenance(PdM)
Definition
A maintenance strategy that uses condition monitoring data, analytics, and failure prediction models to determine the optimal timing for maintenance intervention. Predictive maintenance aims to perform maintenance just before failure is expected, maximizing equipment life while minimizing unplanned downtime. It goes beyond condition monitoring by adding prediction — estimating the remaining useful life based on degradation trends.
Context
The most sophisticated maintenance strategy, combining condition monitoring data with statistical models (Weibull, degradation models) to predict when failure will occur. In Reliatic, predictive maintenance is supported by reliability analytics (Weibull analysis, trend monitoring) and condition indicator tracking.
Examples
- A degradation model fitted to bearing vibration data predicts functional failure in 45 days with 90% confidence, triggering a planned replacement during the next available production window.
- Predictive maintenance of a heat exchanger uses corrosion rate trending to project when the wall thickness will reach retirement thickness, scheduling tube bundle replacement at the optimal turnaround.
Applicable Standards
- ISO 13381-1
- ISO 17359:2018
- IEC 60300-3-11
Related Terms
The process of monitoring a parameter of condition in equipment (vibration, temperature, pressure, flow, oil quality, acoustic emission, electrical properties) to identify significant changes indicative of a developing failure. Condition monitoring is the technical enabler of predictive maintenance, detecting degradation at the P point on the P-F curve to allow planned intervention before functional failure.
Preventive Maintenance(PM)Maintenance performed at predetermined intervals or according to prescribed criteria to reduce the probability of failure or functional degradation. Preventive maintenance includes time-based activities (replacement, overhaul, calibration at fixed intervals), usage-based activities (at fixed operating hours or cycles), and condition-based activities (when monitored parameters indicate intervention is needed). The goal is to prevent failures that would otherwise occur.
Run-to-Failure(RTF)A deliberate maintenance strategy where no preventive or predictive maintenance is performed, and the item is allowed to operate until it fails, at which point it is repaired or replaced. RTF is a valid strategy for non-critical items where the consequence of failure is low, the item is not safety-related, redundancy exists, or the failure mode is random (no wear-out pattern). RTF is not the same as neglect — it is a conscious, risk-based decision.
Remaining Useful Life(RUL)The estimated time from the present until an equipment item reaches a condition where it can no longer perform its intended function safely and must be repaired, replaced, or retired. For static equipment, RUL is typically calculated from the current measured wall thickness, the minimum required thickness, and the corrosion rate. For rotating equipment, RUL may be estimated from degradation models fitted to condition monitoring data.
P-F IntervalThe time between the point at which a potential failure is first detectable (P — the onset of detectable degradation) and the point of functional failure (F — the point at which the item can no longer perform its required function). The P-F interval determines the maximum inspection interval for condition-based maintenance — inspections must occur at intervals shorter than the P-F interval to detect the developing failure before functional failure occurs.
Where this appears in Reliatic
This concept is used in the Reliability Analytics module.