This free FMEA checklist is for quality, design and manufacturing engineers who run a failure mode and effects analysis on a new or changed process or design, and then have to show the actions were done. One run is one FMEA: a process FMEA (PFMEA) for one line or operation, or a design FMEA (DFMEA) for one product or system. It plans the scope, the team and the rating scales, gathers the inputs, and lists the functions or process steps. Then it identifies failure modes, effects, causes and current controls, rates severity, occurrence and detection, and prioritises the risks by RPN or an action priority table. The quality lead approves the action plan before work starts. The last phase implements the actions, re-rates the lines and keeps the FMEA live. The inputs match the FMEA type, and a mandatory action task appears when any effect reaches the top of the severity scale.
A failure mode and effects analysis (FMEA) is a team method for finding how a process or a product could fail before it does. For each process step or design element the team asks how it could fail, what the effect would be, why it would happen and what already prevents or catches it, then rates the risk and acts on the worst. The Institute for Healthcare Improvement describes it as a systematic, proactive method, particularly useful for a new process before it starts and for a proposed change to an existing one. IEC 60812, the general international standard, covers how an FMEA is planned, performed, documented and maintained, and the AIAG & VDA FMEA Handbook is the automotive industry’s reference manual for design and process FMEAs. This checklist follows the common steps without reproducing either: use the rating scales your procedure or your customer sets.
An FMEA looks forward. A root cause analysis (8D) looks back at a failure that has already happened, and its fix should end with an updated FMEA. A CAPA is the formal loop that records a nonconformity and proves the action worked. When a change to a released design or process is proposed, the manufacturing change order asks for the FMEA to be reviewed, and the manufacturing quality control checklist runs the inspection points the FMEA and control plan decide on.
Analyses: a system, subsystem or part, and its interfaces with the user and its surroundings.
Inputs: a block or boundary diagram, requirements, the design verification plan and past field failures.
Typical actions: a design change, a different material or tolerance, or a new verification test.
Analyses: each operation from receiving material to despatch, including handling, rework and inspection.
Inputs: a process flow diagram, product requirements and special characteristics, the control plan and past scrap and complaints.
Typical actions: mistake-proofing, a better prevention control, then a better detection control.
Asks: why did this failure happen, and why was it not caught?
When: after a complaint, a repeat defect or a failure that escaped.
Feeds back: the new failure mode and revised ratings go into the FMEA, so the next design or process starts with them.
Eight phases take one FMEA from scope to review. Phase 2 shows the inputs for a process or a design FMEA, Phase 6 shows the RPN or the action priority task to match the method you choose, and a mandatory action task appears when any effect reaches the top of the severity scale. The quality lead approves the action plan before any action starts.
An FMEA with a vague scope turns into a list of everything that could ever go wrong. Fix the boundaries, the team and the rating scales before the first session, so every line is rated the same way.
Shows the process flow diagram and the control plan for a process FMEA, or the block diagram and the design verification plan for a design FMEA. Past failures are the best evidence the team has: bring them to the first session.
A failure mode is a way a function fails, so the functions come first. Write each one as a verb and a noun with a measurable requirement, and the failure modes almost write themselves.
Work one function at a time: how can it fail, what happens if it does, and why would it fail. Then record what already prevents the cause and what would catch it.
Rate on the scales agreed in Phase 1, as a team, with evidence beside each number. A rating nobody can explain will not survive the first review.
Shows the RPN or the action priority task to match the method agreed in Phase 1, and a mandatory action task when any effect reaches the top of the severity scale. Severity comes first: a low score elsewhere does not make a safety effect acceptable.
The best actions lower severity or occurrence by changing the design or the process. Extra inspection lowers only the detection rating, so it is the last resort. The quality lead approves the plan before any action starts.
An FMEA is finished only when the actions are in place and the lines are re-rated on evidence. After that it stays open as a living document, reviewed whenever the design or process changes.
Every FMEA line carries three ratings, and the team turns them into a priority. The scales themselves belong to your procedure or your customer: write a definition for each number before the first session, and keep them for the whole FMEA. The table sets out what each rating asks, so the team rates the same thing in the same way.
| Rating or method | The question it answers | Rated for | What lowers it | Common mistake |
|---|---|---|---|---|
| Severity | How serious is the worst effect if this failure mode happens? | Each failure mode, using its most serious effect | Only a change to the design or process that removes or reduces the effect | Lowering severity because a new inspection was added |
| Occurrence | How likely is this cause, with today’s prevention controls? | Each cause | Removing the cause, mistake-proofing, or a better prevention control | Rating from memory when scrap, test or field data exists |
| Detection | How likely are today’s controls to catch the cause or the failure mode before it leaves? | Each cause or failure mode, with its detection controls | A better test, gauge or inspection, or detection moved closer to the cause | Counting a control that is in the procedure but not done on the shift |
| Risk priority number (RPN) | Which lines should we work on first? | Each line: severity × occurrence × detection | Lowering any of the three ratings | Using one RPN cut-off and missing a high severity with a low score |
| Action priority | Does this line need action: high, medium or low? | Each line, from a lookup table that weighs severity first, then occurrence, then detection | The same actions, judged in that order | Using a table nobody agreed, or changing it after the ratings are in |
Why severity comes first. On three 1 to 10 scales the RPN runs from 1 to 1,000, and very different risks can share a number: a severity of 9 with an occurrence of 2 and a detection of 5 gives 90, and so does a severity of 2 with an occurrence of 9. That is why it is worth adding a severity rule, such as an action for every effect at the top of the scale whatever its other ratings, and why action priority tables weigh severity before the other two.
Which actions count most. A design or process change that removes the cause lowers occurrence for good. Mistake-proofing stops the cause or catches it at once. Extra inspection lowers only the detection rating and costs money on every part, so put it last. Re-rate only on evidence that the action works.
The first task asks which kind of FMEA this is. A process FMEA shows the flow diagram and control plan tasks, a design FMEA shows the block diagram and design verification plan. The method chosen in Phase 1 shows the RPN or the action priority task, set up in the template designer with no extra templates to keep in step.
The quality lead is chosen on the first task. The plan approval is assigned to them alone, and nothing after the approval step can be completed until they record a decision. When an effect reaches the top of the severity scale, the approval also asks whether every one of those lines has an action.
The last tasks record the review triggers and the next review date. A recurring schedule can start the periodic review for each line or product, and the Reports Grid shows every FMEA in progress and which tasks are overdue. The worksheet, the evidence and each decision sit on the task that produced them.
CheckFlow does not hold the FMEA worksheet or score it for you. Keep the worksheet in your FMEA software or spreadsheet and attach it to the tasks. CheckFlow runs the work around it: who does each step, by when, with what evidence, and who approved the plan. CheckFlow for manufacturing shows how plants run their other quality and production checks the same way.
When a failure does happen, investigate it with the Root Cause Analysis (8D) Checklist and log it in the CAPA Checklist, then bring the new failure mode back into this FMEA. A design change goes through the Manufacturing Change Order Process Checklist, and a software feature reaches customers through the Feature Release Process Checklist, which can carry a design FMEA’s actions to release.
Root Cause Analysis (8D) Checklist
CAPA (Corrective & Preventive Action) Checklist
Manufacturing Change Order Process Checklist
Manufacturing Quality Control Checklist
Process Standardization Checklist
Describe the process in a sentence and the AI checklist generator drafts the phases and tasks for you.
Failure mode and effects analysis is a structured team review of a process or a product, done before failures happen. For each process step or design element the team lists the ways it could fail, the effect of each failure, its causes and the controls already in place. It rates severity, occurrence and detection, decides which risks need action, then checks the actions worked. The Institute for Healthcare Improvement calls it proactive: it corrects a process before an adverse event rather than after one.
A design FMEA (DFMEA) asks whether the design could fail to do its job: the parts, materials, tolerances and interfaces. A process FMEA (PFMEA) asks whether the manufacturing or service process could fail to deliver a design that is right, operation by operation. The method is the same. The inputs differ: a block diagram and a design verification plan for a DFMEA, a process flow diagram and a control plan for a PFMEA. The first task of this checklist asks which one you are running and shows the matching inputs.
Multiply the three ratings: severity × occurrence × detection. On 1 to 10 scales the result runs from 1 to 1,000, and the team works on the highest numbers first. Use it with care. Very different risks can share a number, so a severe effect with a low occurrence can rank below a trivial one that happens often. So add a severity rule as well, such as an action for every effect at the top of the scale whatever its RPN.
Action priority is a way to rank FMEA lines without multiplying the ratings. The team looks each line up in a table that weighs severity first, then occurrence, then detection, and gets a high, medium or low priority. A high severity with a weak prevention control comes out high even if detection is good. Use the table your procedure or your customer sets. This checklist supports either method, or both, and shows the matching task in Phase 6.
Treat the FMEA as a living document. Review it when the design or process changes, when a new material, supplier, machine or location is introduced, when the product is put to a new use, and when a complaint, field failure or repeat defect shows a failure mode the FMEA missed or rated too low. Set a periodic review date as well, so it is looked at even when nothing seems to have changed. The last phase of this checklist records the triggers and the next review date.
People who know the design, the process and its problems: design and process engineers, quality, production, maintenance and at least one operator who does the work. For a design FMEA add test and service people, and for a process FMEA someone from the next operation or the customer’s plant if you can. Four to eight people with a facilitator who keeps to the method is a workable size. One person filling in a worksheet alone misses the failures nobody has written down.
FMEA looks forward: it asks how a process or product could fail and reduces the risk before it reaches a customer. Root cause analysis looks back at a failure that has already happened and finds why it occurred and why it was not caught. The two feed each other. A closed 8D investigation should add its failure mode and revised ratings to the FMEA, and an FMEA review should use past 8D and CAPA records as evidence.
14-day free trial, no card required. The Business plan is $10 per user per month after the trial. Full details at checkflow.io/pricing.
14-day free trial, no credit cardFull refund in your first 30 daysUS or EU hosting
