What FMEA Actually Is
Failure Mode and Effects Analysis is a structured analytical technique that asks three questions about every potential failure mode in a system, process, or product.
How can this fail?
What happens when it fails?
What can be done about it?
The technique was developed by the US military in the 1940s for reliability analysis of weapons systems, adopted by NASA for spacecraft design in the 1960s, and brought into automotive manufacturing by Ford in the 1970s.
Today it is a formal requirement for automotive suppliers under IATF 16949, a standard analytical tool in pharmaceutical manufacturing under FDA guidelines, and a widely used reliability engineering practice across aerospace, food, medical device, and general manufacturing industries.
The fundamental value of FMEA is that it forces a systematic examination of potential failures before they occur — rather than the reactive analysis that most organizations conduct after failures have already caused downtime, quality losses, or safety incidents.
A team sitting down to conduct an FMEA is asking: given everything we know about how this equipment or process works, what could go wrong, how bad would it be, and what are we going to do about it?
The answers to those questions produce a prioritized action list that directs maintenance investment, process improvement, and design change toward the failure modes that matter most.
The Two Types of FMEA Most Relevant in Manufacturing
FMEA exists in several variants designed for different applications.
Two are most relevant for manufacturing operations teams.
Design FMEA (DFMEA)
Design FMEA is applied to the design of a product or piece of equipment — examining potential failure modes that could be designed into the product or asset and addressing them before manufacture or installation.
In a manufacturing context, DFMEA is most relevant when specifying new production equipment — identifying potential failure modes in the equipment design that could be addressed through specification changes, component selection, or design modifications before the equipment is installed and operating.
It is also used by product engineering teams to identify potential product failure modes that manufacturing processes need to prevent or detect.
Process FMEA (PFMEA)
Process FMEA is applied to manufacturing processes — examining how each step in a production or maintenance process can fail, what the effect of each failure would be on the output, and what controls exist to prevent or detect the failure.
In manufacturing maintenance, PFMEA is most directly relevant when analyzing maintenance processes — identifying how a PM task can be incorrectly executed, what the consequences of that incorrect execution are, and what controls prevent or detect the error.
For production operations teams, PFMEA identifies how each production process step can produce nonconforming output — making it a quality management tool as well as a reliability tool.
How FMEA Works: The Core Process
An FMEA analysis is conducted by a cross-functional team — typically including maintenance engineers, quality engineers, production supervisors, and operators who have direct knowledge of the equipment or process being analyzed.
The analysis is documented in an FMEA worksheet that captures the following elements for each potential failure mode.
Function
What is the item supposed to do? What is its required function and performance standard?
Potential Failure Mode
In what specific way could the item fail to perform its required function?
A filling machine filling head has the function of dispensing a precise volume of product into each container.
Potential failure modes include: dispenses too much product, dispenses too little product, fails to dispense entirely, dispenses intermittently.
Each failure mode is listed separately because each has different causes, different effects, and different appropriate responses.
Potential Effect of Failure
What happens when this failure mode occurs?
Effects are described in terms of what the customer or downstream process experiences — not in terms of what the equipment does internally.
For the filling machine overfill failure mode, the effect might be: overweight products that fail checkweigher inspection, customer complaints, regulatory non-compliance in weight-controlled products.
Severity (S)
How severe is the effect of the failure?
Severity is rated on a scale of 1 to 10 — where 1 means the failure has no noticeable effect and 10 means the failure causes a safety incident or regulatory violation without warning.
Severity is a property of the failure effect, not the failure mode itself. The same failure mode producing different effects in different contexts has different severity scores.
Potential Causes of Failure
What specific causes could produce this failure mode?
For the filling machine overfill failure mode, potential causes include: worn fill nozzle seal allowing dribble, fill head timing calibration drift, product viscosity variation causing flow rate change.
Each cause is listed separately because each has different detectability and different appropriate preventive measures.
Occurrence (O)
How often is this failure mode likely to occur given the potential cause and any existing prevention controls?
Occurrence is rated on a scale of 1 to 10 — where 1 means failure is extremely unlikely and 10 means failure is almost certain to occur without intervention.
Current Process Controls
What controls currently exist to either prevent this failure mode from occurring or to detect it if it does occur?
Prevention controls reduce the occurrence rating. Detection controls affect the detection rating but do not change the severity or occurrence.
Detection (D)
How likely is the current control to detect the failure mode before it reaches the customer or causes significant consequence?
Detection is rated on a scale of 1 to 10 — where 1 means the control will almost certainly detect the failure and 10 means detection is nearly impossible with current controls.
Note that a lower detection score means better detection — this is counterintuitive but reflects the FMEA convention where lower numbers are always better across all three rating dimensions.
Risk Priority Number (RPN)
The RPN is calculated by multiplying the three scores together.
RPN equals Severity multiplied by Occurrence multiplied by Detection.
The maximum possible RPN is 1,000 (10 × 10 × 10).
High RPN values indicate failure modes that warrant priority attention — either because they are severe, common, hard to detect, or some combination of all three.
Recommended Actions
What specific actions should be taken to reduce the RPN for high-priority failure modes?
Actions can target any of the three RPN components. Reducing severity requires design change or operating procedure modification. Reducing occurrence requires better prevention controls or maintenance interventions. Improving detection requires adding or improving monitoring, inspection, or testing.
Responsible Person and Target Completion Date
Who is responsible for implementing each recommended action, and by when?
Without this assignment, the FMEA analysis produces insights that never become improvements.
How to Use RPN Scores Effectively
The RPN score is a useful prioritization tool but it has limitations that FMEA practitioners consistently encounter.
Two failure modes with the same RPN can represent very different risk profiles.
A failure mode with severity 10, occurrence 1, and detection 1 has an RPN of 10.
A failure mode with severity 1, occurrence 10, and detection 1 also has an RPN of 10.
These two failure modes require completely different responses — the first is a low-probability catastrophic event requiring a prevention-focused response, the second is a high-frequency trivial event that may require nothing at all.
Experienced FMEA practitioners use RPN scores as a starting point for prioritization rather than as the final arbiter of action.
They pay particular attention to any failure mode with a severity score of 9 or 10 regardless of its RPN — because safety and regulatory consequences warrant priority attention even if occurrence and detection scores are favorable.
They also look at the individual component scores rather than the composite RPN — a high detection score (meaning poor detection) combined with moderate severity and occurrence may warrant detection improvement even if the RPN is not the highest on the list.
FMEA Outputs and Maintenance Program Design
The most direct value of FMEA for manufacturing maintenance programs is the structured list of prioritized failure modes it produces — and the specific maintenance task recommendations that follow from the analysis.
Each high-priority failure mode in the FMEA produces one of three maintenance recommendations.
Condition monitoring task
When the failure mode produces a detectable precursor signal within a useful P-F interval, the FMEA recommends condition monitoring as the technically feasible control.
The FMEA identifies what to monitor, what threshold indicates the P point has been reached, and what response is required when the threshold is crossed.
This output feeds directly into condition-based maintenance program configuration — specifying which condition signals to monitor on which assets and what maintenance actions to trigger when configured thresholds are crossed.
Time-based or usage-based PM task
When the failure mode shows consistent age-related or usage-related deterioration but no reliably detectable precursor signal, the FMEA recommends a scheduled PM task with an interval calibrated to the failure mode's characteristic age or usage.
Design or procedure modification
When neither condition monitoring nor PM is technically feasible or cost-effective for a high-consequence failure mode, the FMEA recommends a design change or operating procedure modification that either eliminates the failure mode or reduces its consequence to an acceptable level.
FMEA in Regulated Manufacturing
FMEA is a formal requirement in several regulated manufacturing industries — and understanding these requirements helps operations teams prioritize where to focus their FMEA investment.
Automotive manufacturing under IATF 16949
The AIAG-VDA FMEA methodology is the current standard for automotive suppliers — covering DFMEA, PFMEA, and Monitoring and System Response (MSR) analysis.
FMEA documentation is a customer-auditable deliverable in most automotive supply chain relationships.
Pharmaceutical manufacturing under FDA guidelines
Process FMEA is a recognized tool for pharmaceutical process risk assessment under the FDA's Process Analytical Technology (PAT) framework and the ICH Q9 guideline on quality risk management.
Medical device manufacturing under ISO 14971
Risk analysis using FMEA methodology is a formal requirement of the medical device risk management standard ISO 14971.
Frequently Asked Questions
How often should an FMEA be updated?
An FMEA is a living document that should be reviewed and updated when equipment or process changes occur, when new failure modes are identified from actual failure events, and when recommended actions are completed and their effectiveness needs to be reflected in updated occurrence and detection scores.
A static FMEA that was completed once and never updated becomes progressively less useful as the equipment and operating context evolve.
Who should participate in an FMEA analysis?
The most effective FMEA teams are cross-functional — combining maintenance engineers who understand failure mechanisms, quality engineers who understand the product and process requirements, production supervisors who understand the operational context, and operators who have direct daily experience with the equipment being analyzed.
External facilitators with FMEA methodology expertise are valuable for teams conducting their first analysis — both for methodology guidance and for avoiding the institutional bias that makes it difficult to identify failure modes that have never occurred but could.
What is the difference between FMEA and root cause analysis?
FMEA is a proactive analysis conducted before failures occur — identifying potential failure modes and taking preventive action.
Root cause analysis is a reactive analysis conducted after a failure has occurred — identifying the specific chain of causes that produced the failure and taking corrective action to prevent recurrence.
The two techniques are complementary. Root cause analysis findings should feed back into the FMEA — adding new failure modes that were not anticipated in the original analysis and updating occurrence and detection scores based on real failure evidence.
FMEA produces a prioritized list of what can go wrong, how bad it would be, and what to do about it. The maintenance program that follows from that analysis is focused on the failures that actually matter rather than the failures that are most visible or most recent.
Login
Blog
