OptimalAvailability Studio™Wiki

OptimalAvailability Studio™ Wiki / RCM to JA1011

Method

Reliability-centred maintenance

RCM decides what must be done so an asset keeps doing what its users need, in its actual operating context. It is not a maintenance task list, it is the logic that produces one, and SAE JA1011 sets the minimum criteria a process must meet to earn the name.

What RCM is

Reliability-centred maintenance is a structured process for determining the maintenance requirements of a physical asset in its operating context. SAE JA1011 (1999) defines the seven questions any process must answer to be called RCM; SAE JA1012 (2002) is the guide that amplifies it. The best-known development is RCM II, from John Moubray. The output is a task for every failure mode that is worth managing, and a deliberate decision to run others to failure.

The seven questions

To JA1011, RCM answers these in order, for the asset in its context:

1. FunctionsWhat the asset must do, and to what performance standard, in this operating context.
2. Functional failuresThe ways it can fail to deliver that standard.
3. Failure modesWhat can cause each functional failure, at enough detail to manage it.
4. Failure effectsWhat happens when each failure mode occurs: evidence, damage, downtime.
5. Failure consequencesHow and how much each failure matters: safety, environment, operations, cost.
6. Proactive tasksThe task and interval that will predict or prevent each failure, where one is worth doing.
7. Default actionsWhat to do where no proactive task is suitable: failure-finding, redesign or run-to-failure.

Failure consequences drive the choice

RCM sorts every failure mode by consequence before choosing a task, because the consequence, not the technical interest of the failure, decides what is worth doing:

Hidden or evidentFirst, is the failure evident to the operating crew on its own, or hidden until a second failure or a check reveals it. Hidden failures need failure-finding.
Safety and environmentalA failure that could hurt someone or breach the environment. If no proactive task reduces the risk to tolerable, redesign is mandatory.
OperationalA failure that costs production, throughput or quality, beyond the repair cost. Worth a task when the task costs less than the consequence.
Non-operationalA failure whose only cost is the repair. A task is justified only if it is cheaper than the repair it avoids.

Task selection

Proactive tasks, chosen only when technically feasible and worth doing:

On-condition (predictive): detect a potential failure and act inside the P-F interval. Preferred where a P-F interval exists.
Scheduled restoration: overhaul at a fixed age, where a clear wear-out life exists.
Scheduled discard: replace at a fixed age regardless of condition, same requirement.

Default actions, where no proactive task suits:

Failure-finding: periodically test a hidden function (for example a protective device).
Redesign: mandatory for intolerable safety or environmental risk with no effective task.
Run-to-failure: a legitimate, deliberate choice where the consequence is minor.

On-condition and the P-F interval

An on-condition task is only feasible when the failure gives warning: a detectable potential failure with a P-F interval that is reasonably consistent and long enough to act within. The monitoring interval must be no more than half the P-F interval so an inspection reliably falls inside the window. This is the exact hand-off to OptimalTREND™, which works the predictive domain and lengthens that interval by detecting earlier. The DIPF curve and the interval rule are covered in depth in the OptimalTREND™ wiki.

Where OptimalAvailability Studio™ fits

OptimalAvailability Studio™ runs the RCM analysis as a living model: functions and failure modes drawn from the FMECA, consequences classified, tasks and intervals selected against the JA1011 logic, and the result packaged as a maintenance strategy that feeds the CMMS, the on-condition tasks to OptimalTREND™, and the criticality to OptimalSPARES™.