Condition-Based Maintenance: How to Act on Real Asset Health Signals

Condition-based maintenance means maintaining equipment based on actual condition, not only fixed calendar intervals.

Instead of servicing an asset every month because the calendar says so, the team uses signals such as vibration, temperature, pressure, running hours, oil condition, energy use, leakage, noise, or inspection findings to decide when work is needed.

This can reduce unnecessary work and help catch failures early. But condition-based maintenance only works when the team knows how to act on signals.

How condition-based maintenance differs from time-based PM

Time-based maintenance uses fixed intervals:

• Every week
• Every month
• Every quarter
• Every 6 months

This works well for compliance routines, safety checks, cleaning, lubrication, and predictable wear.

Condition-based maintenance uses asset behavior:

• Bearing temperature is rising
• Vibration exceeds baseline
• Pressure drop is increasing
• Oil analysis shows contamination
• Motor current is abnormal
• Compressor energy use is rising
• Inspection shows early wear

The decision is based on condition, not only time.

What signals plants can monitor

Condition-based maintenance does not always require advanced sensors. Many useful signals can come from inspections, readings, and technician observations.

Common condition signals include:

• Vibration level
• Bearing temperature
• Motor current
• Pressure
• Flow
• Oil condition
• Filter differential pressure
• Noise
• Leakage
• Energy consumption
• Running hours
• Cycle count
• Visual wear

A practical plant starts with signals that are easy to capture and strongly connected to failure risk.

The biggest mistake: collecting data without action

Many plants collect readings but still suffer breakdowns. The problem is usually not data collection. The problem is follow-up.

If a vibration reading is abnormal, what happens next?

• Who reviews it?
• What threshold matters?
• Is a work order created?
• Is the asset critical?
• Is the repair urgent or plannable?
• Are spare parts available?
• Was the issue verified after repair?

Without this workflow, condition monitoring becomes another report.

How to set practical alarm rules

Condition-based maintenance needs clear rules so the team does not ignore signals or overreact.

A simple rule model:

Normal: Continue routine monitoring.

Watch: Reading is moving away from baseline. Review trend and operating condition.

Plan: Reading crosses action threshold. Create planned inspection or corrective work.

Urgent: Reading indicates high risk. Prioritize immediate action.

The same value may mean different things depending on asset criticality. A minor fan and a critical production compressor should not have the same response rules.

Where work orders fit

When condition data shows risk, it should create maintenance action.

A practical workflow:

1. Reading or inspection detects abnormal condition.
2. Supervisor reviews severity and asset criticality.
3. Work order is created.
4. Technician inspects or repairs the asset.
5. Spare parts are planned or consumed.
6. Reading is checked again after the work.
7. Asset history is updated.

This is where work order management software connects condition signals to execution.

Where preventive maintenance fits

Condition-based maintenance does not remove preventive maintenance. It improves it.

For example:

• A PM checklist can include condition readings
• A meter reading can trigger service after actual usage
• A recurring inspection can detect wear trends
• A failure pattern can change PM frequency
• A condition alert can create follow-up work

A preventive maintenance software setup should allow teams to combine calendar-based PM, meter-based PM, inspections, and condition-based actions.

How to avoid data noise

Condition monitoring can create too many signals. If everything becomes urgent, nothing is urgent.

To reduce noise:

• Start with critical assets
• Define baseline readings
• Use clear thresholds
• Separate warning from urgent action
• Review repeated nuisance alerts
• Consider operating conditions
• Avoid creating work orders for every minor fluctuation
• Track whether alerts actually prevented failures

The goal is not more alerts. The goal is better decisions.

What to track

Useful CBM metrics include:

• Number of abnormal readings
• Alerts converted to work orders
• Time from signal to action
• Failures detected early
• Emergency breakdowns avoided
• Repeat alerts by asset
• Condition-based work completed
• PM interval changes based on condition

A maintenance analytics software report can show whether condition-based maintenance is reducing downtime or only creating more work.

Bottom line

Condition-based maintenance works when condition signals become clear decisions.

Start with critical assets, simple readings, clear thresholds, and disciplined follow-up. Then connect abnormal findings to work orders, PM updates, spare readiness, and asset history.

Data alone does not improve reliability. Action on the right data does.