Condition-based vs preventive maintenance: how to choose a strategy by failure consequence

Start from what the asset must do

The international standard on asset management frames the job as coordinating decisions so that the asset portfolio delivers the organisation's objectives over its lifecycle. Its practical implication is to begin from criticality and business value, not from a generic task list: a maintenance plan that cannot name the outcome it protects is unanchored.

Pick tasks by failure consequence

Reliability-centred maintenance selects tasks by asking what the asset must do, how it can fail, and what each failure would cause. Where no preventive task is justified, it uses a deliberate default action instead of an arbitrary interval. Applying the same interval to every asset wastes effort on low-consequence failures and can miss hidden ones. Consequence, not habit, should choose the task.

This needs consistent failure history: the same definitions of a failure and a repair across teams, otherwise the data cannot tell you which assets fail and why.

Condition monitoring has to earn its place

Condition-based approaches rely on measurement that separates ordinary variation from a real signal. Statistical process control does this with time-ordered charts whose limits come from a stable period, and it depends on trustworthy measurement systems first. A chart that reacts to every wobble is worse than none, and a signal says when something changed, not why.

That is why the playbook asks you to justify monitoring with data: does the failure give a detectable warning early enough to act, and does the cost of monitoring beat the cost of the failure it prevents?

Maintenance is one part of a system of practices

In a sample of manufacturing plants, lean practices clustered into interdependent bundles that included total productive maintenance, and together explained roughly 23% of the variation in operational performance, with plant size influencing how they were implemented. Read this as a reason to fit maintenance to the rest of operations and to your context, not to copy one tool.

The path, step by step

These steps come from the Maintenance and asset reliability playbook. Each has a checkpoint you can observe.

1. 01

   Classify assets by function and criticality

   List the assets that carry operational, safety or environmental consequence. For each, state the function it must perform and what happens when it fails.

   CheckpointA criticality ranking tied to operational and safety consequences, with the top assets named.

   Common failureTreating every asset as equally critical and spreading maintenance thin.

2. 02

   Capture failure history with consistent definitions

   Record failure mode, cause, detection, downtime and repair for critical assets. Use the same definitions across shifts and sites so the history can be compared.

   CheckpointA failure log with consistent categories and enough history to rank modes.

   Common failureChoosing maintenance strategy from vendor claims because the failure history is unusable.

3. 03

   Choose tasks by failure consequence

   For each significant failure mode, decide whether a scheduled task, condition-based check, redesign or run-to-failure is justified. Delete tasks that cannot address a credible mode.

   CheckpointA task decision per major failure mode, including the explicit run-to-failure decisions.

   Common failureAdding more preventive tasks to every asset and calling it reliability.

4. 05

   Justify condition monitoring with data

   Use condition-based or predictive monitoring only where the failure mode gives a usable warning and the consequence justifies the cost. Pilot on one asset class and compare with the existing task.

   CheckpointA pilot showing earlier detection or lower cost than the existing task, with false alarms counted.

   Common failureBuying sensors for assets whose failures give no warning or carry little consequence.

What this guide does not settle

These sources are a standard, a practitioner method, a textbook and a cross-sectional plant survey. None of them proves that condition-based beats preventive maintenance in a given plant, and none estimates the financial return of a maintenance programme on its own.