What is predictive maintenance?
Using condition data and trends to forecast when equipment needs attention — so work is done before failure, but not before it is needed.
Predictive maintenance uses the measured condition of equipment, and how it is changing over time, to forecast when it will reach an unacceptable state. Work is then planned for the right moment: after the trend shows a problem, and before the equipment fails.
It sits between waiting for a failure and servicing everything on a fixed timetable. It needs regular, reliable readings and a plan for acting on the forecast. A forecast is only an estimate, so poor data or a forecast that is not checked can bring missed failures or needless work, and required inspections must continue.
- Monitor. Take condition readings regularly and in a consistent way, by hand or from sensors.
- Trend. Chart each reading over time against its normal range and alert level.
- Forecast. Estimate when the reading will reach the alert or failure level.
- Plan. Choose the work, the date inside the window, the spares and the access needed.
- Act. Repair or replace before failure, at a time that suits airfield operations.
- Verify. Check the result, compare the forecast with what happened, and refine the method.
Work is done on a timetable whatever the condition. Reliable and simple, but some items are serviced too early and others fail between visits.
Work follows measured condition and forecast. It goes where it is needed and is timed before failure, but depends on good data and sound judgement.
Frequently Asked Questions
Q1What is predictive maintenance in AGL?
Predictive maintenance uses the measured condition of equipment, and how it is changing, to forecast when it will reach an unacceptable state. Repair or replacement is then planned for the right time, before the equipment fails and without replacing it long before it is needed.
Q2How is it different from preventive and corrective maintenance?
Corrective maintenance acts after a failure. Preventive maintenance acts at fixed intervals, whatever the condition. Predictive maintenance acts according to condition and forecast, so work goes where it is needed and is done before failure.
Q3What data does it use?
Readings that change as equipment ages or degrades, such as insulation resistance of circuits and cables, regulator current and temperature, transformer and connector checks, light output of fixtures, and the history of faults and repairs. Readings must be taken regularly and in a consistent way.
Q4Does predictive maintenance need AI?
Not necessarily. Many sites start with regular readings and simple trend charts against alert levels. AI and other analysis can help when many readings must be considered together, but good data and a clear plan for acting on the forecast matter more than the tool.
Q5Does it replace scheduled inspection and maintenance?
No. Required inspections, tests and serviceability checks continue, and the lighting must still meet the levels the aerodrome and regulator require. Predictive maintenance adds condition information so that work can be better timed and targeted.
Q6What should be in place before starting?
A baseline of good readings, agreed alert levels, regular and reliable data collection, clear responsibility for acting on a forecast, spares and lead times that fit the forecast window, and a way to check afterwards whether the forecast was right.
See It. Understand It. Then Learn How to Engineer It.
Recognising an AGL system visually is the first step. The next step is knowing how to design, install, test, commission and maintain them correctly.
Build Practical AGL Engineering Skills
Go deeper with structured training from Precision AGL Academy, covering:
Learn how each light, sign and circuit works as part of a complete airport lighting system, not just as an isolated diagram.