What Is Condition-Based Maintenance?

Predictive maintenance takes the analysis further by estimating future condition or failure using trends and models. A simple threshold that triggers a filter change is condition-based; it does not require a prediction algorithm.

These approaches can work together. A system can retain scheduled safety tasks while condition findings determine additional cleaning, adjustment or component replacement.

Start with the failure you want to detect and whether deterioration gives a meaningful warning. Examples include rising vibration in rotating equipment, filter pressure drop, unusual temperature or visible wear. The measurement must suit the asset and the failure mechanism.

Not every reading indicates a defect. A temperature or vibration change can reflect different operating load. Capture the conditions alongside the measurement and compare like with like. An unexplained figure without context is a weak basis for a repair instruction.

Condition monitoring is most useful when there is enough warning to investigate and act. Sudden failures without a detectable lead-up need other controls, such as appropriate planned tasks, standby arrangements or replacement decisions.

1. Establish the baseline. Identify the equipment, normal operating conditions and available history.
2. Select the method. Define the inspection or measurement and the competence required.
3. Set the criteria. Use manufacturer information and suitable technical assessment to define what requires attention.
4. Assign the response. Identify who reviews a finding, the escalation route and the next action.
5. Check the result. Record completed work and whether the condition returned to an acceptable state.

A warning without an owner is unfinished management. Specify what happens if a reading is missing, a sensor fails or an inspection cannot be completed. Those events must not be recorded as evidence of healthy equipment.

Condition evidence does not automatically cancel statutory inspections, manufacturer requirements or other established maintenance duties. Assess any proposed change to the programme against those requirements and the asset's risks.

HSE's maintenance guidance addresses safe planning and competence. Its maintenance procedures guidance, written for major-hazard plant, also illustrates the importance of selecting an appropriate maintenance regime for each item. The method needs to fit the equipment and its operating context.

Compare the effort of inspections or monitoring with the consequences of failure and the ability to act on the findings. Expensive continuous monitoring is unnecessary where a reliable periodic inspection provides enough warning. Conversely, occasional observations are inadequate if deterioration can progress between checks.

For industrial premises, load and production patterns affect interpretation. For offices, findings can inform maintenance of building plant without treating every asset as operationally critical.

No. Structured visual inspections and portable measurements can provide condition evidence. Select the method according to the asset, deterioration pattern and required warning period.

No. Retain applicable scheduled tasks and inspections. Condition evidence helps shape the programme where it provides a sound basis for doing so.

Follow the defined escalation and technical response. Depending on the finding, that means checking the reading, investigating the cause or arranging work; it should not be an unexplained automatic closure or reset.

Review detected defects, useful warning time, actions completed and repeated failures. A large volume of readings is not itself evidence of better maintenance decisions.

Explore Condition-Based Maintenance, Asset Maintenance, Mechanical and Planned Replacement Programmes.

Browse all services, maintenance, projects, sectors and locations. Discuss condition-based maintenance with asset details, failure history and the decisions you want condition information to support.