Fix the failure mode with the greatest combination of consequence and Action Priority, not automatically the one with the highest Risk Priority Number. In a process FMEA, start with severity, use the official Action Priority table, then compare existing controls and residual risk.
A completed process FMEA can leave you with dozens of ratings but no obvious first action. Limited engineering time makes prioritization a practical decision, not a scoring exercise. You need a method that identifies serious exposure, directs resources toward meaningful risk reduction, and produces a rationale you can defend during management reviews and audits.
What Is Process FMEA And Why Is Prioritization Difficult?
Process FMEA identifies how a manufacturing or service process could fail, evaluates the effects and causes, and documents controls intended to prevent or detect each failure. The analysis becomes useful when it directs action toward the risks that deserve attention first.
Process Failure Mode and Effects Analysis (PFMEA) examines potential failures at the process level. A design analysis asks how a product design could fail, but PFMEA examines failures connected to process inputs, equipment, methods, materials, measurement systems, handling, and human tasks. You use it before production begins and continue updating it as the process, controls, and operating experience change.
Prioritization becomes difficult when several failure modes receive similar scores or conflicting rating profiles. One risk may have severe consequences but a low occurrence rating, and another may happen often but create a less serious effect. A simple numerical ranking can place the recurring low-severity problem above the rare high-severity problem. That ranking may direct your team toward scrap reduction while leaving a safety, compliance, or major customer risk insufficiently controlled.
How Do You Score Severity, Occurrence, And Detection?
You rate each failure mode for severity, occurrence, and detection on scales from 1 to 10. The ratings describe different parts of the risk and should not be treated as interchangeable.
Severity measures the seriousness of the failure effect. A rating of 1 represents no effect, and a rating of 10 represents a hazardous condition without warning or regulatory noncompliance under the Automotive Industry Action Group (AIAG) and German Association of the Automotive Industry (VDA) method. Severity belongs to the effect, not the cause or inspection method. Better inspection normally changes detection, not the consequence experienced when the failure reaches its point of impact.
Occurrence measures how likely the cause and resulting failure mode are to happen. A rating of 1 represents an extremely unlikely event, and 10 represents an event considered almost certain or inevitable. Detection measures the ability of current controls to find the cause or failure before the effect occurs or the product moves forward. A low detection rating means the controls are likely to detect the issue, and a rating of 10 means there is no effective detection opportunity.
Is Risk Priority Number Still Used In PFMEA?
Risk Priority Number is still useful as supplementary information, but it should not determine action by itself. Modern process FMEA practice uses Action Priority to interpret combinations of severity, occurrence, and detection more carefully.
The Risk Priority Number (RPN) is calculated by multiplying Severity × Occurrence × Detection. The arithmetic is easy, which explains why teams often sort a spreadsheet from the largest RPN to the smallest. The problem is that very different risk profiles can produce similar totals. Multiplication can also allow a low occurrence or detection score to offset a severe effect.
There is no universal RPN threshold that automatically requires action. A rule stating that every score above 100 needs action, for example, can cause your team to ignore serious risks below the cutoff or spend resources on less serious items above it. The American Society for Quality (ASQ) and AIAG/VDA guidance caution against relying on arbitrary thresholds without reviewing severity. Your organization may use internal decision criteria, but those criteria should supplement the recognized rating method rather than replace engineering judgment.
What Changed With Action Priority?
Action Priority assigns High, Medium, or Low priority through official lookup tables rather than a multiplication formula. The tables examine the exact combination of severity, occurrence, and detection, with added weight given to serious effects.
The AIAG/VDA FMEA Handbook introduced Action Priority (AP) to improve how automotive teams select risks for action. You locate the applicable severity, occurrence, and detection combination in the handbook’s AP tables. You cannot calculate AP by adding scores, multiplying them, or converting an RPN range into a priority category. A spreadsheet or software tool should reproduce the official lookup logic accurately.
An AP High result signals that the team needs focused review and documented action planning. AP Medium still calls for a reasoned evaluation of potential improvements, especially when stronger prevention or detection controls are practical. AP Low does not mean the risk disappears or can never require action. Customer requirements, internal experience, special characteristics, regulations, and control effectiveness may justify action at any AP level.
Which Process FMEA Risk Should You Fix First?
Address the highest-severity exposure first when its Action Priority confirms inadequate prevention or detection. After reviewing AP High items, move to AP Medium items and compare how much residual risk each proposed action can remove.
Begin by screening for failure effects involving safety, regulatory compliance, loss of primary function, or serious customer impact. Check whether the AP table places those items in the High category. Then review whether current controls prevent the cause or merely detect the defect after it occurs. Prevention normally provides stronger risk reduction because it reduces the opportunity for the failure to happen.
If several items share the same AP category, compare their severity ratings and current control strength. Review whether the detection method finds the specific failure, runs at the correct process stage, and produces a defined reaction when the result fails. You should also compare the residual risk expected after each proposed action. Fix the item where action can reduce serious exposure, rather than choosing the easiest task simply because it can be closed quickly.
- Screen by severity: Identify effects that could cause the greatest harm or noncompliance.
- Use the AP table: Confirm whether each rating combination receives High, Medium, or Low priority.
- Review controls: Separate genuine prevention from downstream inspection.
- Compare residual risk: Estimate what exposure will remain after the planned action.
- Document the decision: Record why one risk was selected ahead of another.
How Do You Compare Two Failure Modes With Conflicting Scores?
Compare the effect severity first, then use the AP tables to evaluate how occurrence and detection change the action priority. A recurring process defect should not automatically outrank a rare failure with hazardous or noncompliant consequences.
Consider Failure Mode A, which has a severe effect connected to safety or regulatory compliance but is considered unlikely to occur. Failure Mode B happens more often and creates scrap, rework, or a minor customer inconvenience, but its effect is less severe. Failure Mode B could produce the larger RPN because its occurrence rating is higher. Sorting by RPN alone would place it at the top of the action list.
You should look up each full rating combination in the official AP table before making the decision. If Failure Mode A receives AP High and Failure Mode B receives AP Medium, Failure Mode A takes priority, regardless of which item has the larger RPN. The action for Failure Mode A could focus on error-proofing, process prevention, automatic shutdown, or an earlier control point, depending on the cause. Failure Mode B still needs review, but its frequent occurrence does not erase the more serious consequence attached to Failure Mode A.
What PFMEA Prioritization Mistakes Should You Avoid?
The most common mistake is treating the largest RPN as an automatic instruction to act. Other errors include calculating AP with a homemade formula, changing ratings to lower the apparent risk, and closing actions without re-rating the failure mode.
Do not lower severity merely because you installed a better inspection. Inspection changes your ability to detect the failure, but it does not reduce the effect if the failure escapes. Severity should change only when the effect itself changes through a process or product modification. Keeping that distinction clear prevents your team from overstating the value of downstream detection.
Avoid treating the PFMEA as a document created once for approval and then stored away. Process changes, tooling revisions, new failure information, control changes, and corrective actions can make old ratings inaccurate. Teams also lose value when they debate individual rating numbers for hours without deciding what action would reduce risk. Use the rating criteria consistently, document reasonable disagreements, and keep the discussion focused on prevention, detection, ownership, and completion.
How Do You Link PFMEA Actions To The Control Plan?
Transfer every implemented control into the control plan, work instructions, inspection method, or training record that governs daily execution. An action is not closed merely because equipment was purchased or a procedure was drafted.
The PFMEA identifies the failure mode, effect, cause, current controls, recommended action, and revised risk. The control plan defines how the selected process or product characteristic will be controlled during production. It should specify the method, frequency, responsibility, acceptance criteria, and reaction plan needed for consistent execution. The operating instructions must then tell the responsible person how to perform the control and what to do when the result is unacceptable.
After implementation, verify that the action works under actual process conditions. Re-rate severity, occurrence, and detection based on the verified control rather than the expected benefit. Record the residual risk, recalculate RPN when your system retains it, and determine the revised AP from the official table. Update related documents so the PFMEA, control plan, instructions, and training materials describe the same production controls.
How Often Should You Review And Update PFMEA Risks?
Review PFMEA risks whenever new information could change the failure mode, cause, effect, control, or rating. Scheduled reviews are useful, but event-driven updates keep the analysis connected to the real process.
Review the document after process changes, equipment changes, corrective actions, control failures, customer complaints, internal nonconformities, or new operating experience. A rating that was reasonable during launch may no longer represent production performance. New evidence can raise or lower occurrence and detection ratings, but every change needs a documented basis. Avoid changing a score simply to make the risk appear acceptable.
Standards and customer requirements also influence how you maintain the analysis. International Automotive Task Force (IATF) 16949 connects manufacturing process risk analysis with quality planning and control plan development in automotive work. International Organization for Standardization (ISO) 9001 requires risk-based thinking, and PFMEA is a structured method you can use to assess process risk. SAE International J1739:2021 provides guidance for potential failure mode and effects analysis across general manufacturing applications.
How Do You Prioritize Risks In A Process FMEA?
- Score severity, occurrence, and detection from 1 to 10
- Calculate RPN, but do not rely on it alone
- Use the AIAG/VDA Action Priority tables
- Fix high-severity or AP High items first
Make Your First Action Defensible
A useful process FMEA does more than arrange failure modes by score. It helps you identify where serious effects meet weak prevention or detection, then directs action toward meaningful risk reduction. Start with severity, confirm the priority through the official AP tables, and use RPN as supporting information rather than a universal decision rule. After the action, verify its performance, re-rate the risk, and update every connected production document. The risk you fix first should be the one whose treatment leaves your process safer, more controlled, and easier to defend with evidence.
References
- Automotive Industry Action Group: Failure Mode and Effects Analysis
- German Association of the Automotive Industry Quality Management Center: Failure Mode and Effects Analysis
- American Society for Quality: Failure Mode and Effects Analysis
- SAE International J1739: Potential Failure Mode and Effects Analysis
- International Organization for Standardization: ISO 9001 Quality Management Systems
- International Automotive Task Force: IATF 16949.
Nirdosh Jagota is the Consultant & Managing Partner, GRQ Biotech Advisors. Nirdosh Jagota has over 30 years of experience in biotech and pharmaceutical industries, leading innovation in regulatory affairs and quality management.
Dr. Nirdosh Jagota has led global teams across US, Europe, Asia Pacific, Latin America, China, India, and Japan. His expertise has been instrumental in the successful submission, approval, and lifecycle management of biologics, small molecules, and vaccines.
