FMEA is failure mode and effects analysis: a structured way of asking, for each part of a machine or a process, how it could fail, what would happen if it did, and how likely and how detectable that failure is. The output is a ranked list of the failures worth doing something about, and for a maintenance team that something is usually a preventive task on the schedule or an inspection on the checklist. This page gives the meaning in plain terms, the steps as a team runs them, and the difference between an FMEA on a piece of equipment and a process FMEA on a line.
What the three words mean when they are applied to a pump
A failure mode is the way a thing stops doing its job: the seal leaks, the bearing seizes, the impeller wears. The effect is what that does to the operation: the line stops, product is scrapped, a technician is called at night. The analysis is the scoring that follows, where each mode gets a number for how severe the effect is, how often it occurs, and how likely it is to be caught before it matters. Multiply the three and the modes sort themselves into the ones that deserve a task and the ones that do not.
The fmea steps, in the order a maintenance team runs them
First, pick the asset or the process and break it into functions: what it is supposed to do. Second, for each function, list the ways it can fail. Third, for each failure mode, write the effect and the likely cause. Fourth, score severity, occurrence and detection on the scale the team has agreed, and multiply to get the risk priority number. Fifth, for the modes at the top of the list, decide the action: a preventive task, an inspection, a redesign, or a spare on the shelf. Sixth, put the actions where they will be done, which for a maintenance team means the schedule and the checklist, and revisit the scores after the actions have run for a while.
A process fmea against an equipment fmea
An equipment FMEA looks at a machine and its parts. A process FMEA looks at a sequence of operations, a filling line say, and asks how each step can fail to produce what the next step needs: wrong torque, missed inspection, wrong material loaded. The scoring is the same; the failure modes are human and procedural as often as mechanical, and the actions are more often a check or a poka-yoke than a maintenance task. Maintenance teams meet process FMEAs when the plant runs one and the equipment-related modes land on their schedule.
Where the output goes
An FMEA that lives in a spreadsheet after the workshop has changed nothing. The modes that scored high become preventive tasks with an interval on the asset's schedule, or inspection lines on a checklist, and the work orders those tasks raise are the evidence that the analysis is being acted on. When a failure the FMEA predicted happens anyway, the work order history is what says whether the task was done, and the score is revised from what actually happened rather than from the workshop's guess.
Questions people ask about fmea meaning
Is FMEA only for manufacturing?
No. Any operation that owns equipment can run one: a hotel's plant room, a warehouse's dock equipment, a school's boilers. The method does not care what the asset is; it needs people who know how it fails.
What is a risk priority number?
Severity multiplied by occurrence multiplied by detection, each scored on an agreed scale. It exists to sort failure modes, not to be precise; two modes with similar numbers deserve similar attention whatever the arithmetic says.
How often should an FMEA be revisited?
When the equipment changes, when a failure it did not predict happens, and when the actions it produced have run long enough to see whether occurrence fell. A yearly look at the top of the list is the usual habit.
Does a CMMS do the FMEA?
No. People do the analysis; the CMMS holds the tasks it produces and the history that tests them. That history is what makes the second FMEA better than the first.