FMEA Steps Explained (AIAG VDA)

FMEA is a structured, step-by-step process used to systematically identify and mitigate risks in manufacturing and product development. Whether referred to as FMEA steps, PFMEA steps, or the FMEA process, the underlying logic follows a consistent six-step methodology defined by the AIAG-VDA standard.

For a broader introduction to the methodology, see the FMEA guide.

Overview of the FMEA Process

Every Process FMEA follows a six-step logic. Each step builds on the previous one, creating a traceable chain from process definition to risk mitigation:

1. Process Step — Define what is being done
2. Failure Mode — Identify what could go wrong
3. Effect — Determine the impact on the customer or downstream process
4. Cause — Identify why the failure occurs
5. Controls — Define how the failure is prevented or detected
6. Action — Assign priorities and corrective actions

To see this logic applied to a real manufacturing process, view the FMEA example.

Step 1: Define Process Step

The first step is to decompose the manufacturing or operational process into discrete steps. Each step should represent a single operation or activity that transforms an input into an output.

For each step, define:

• The function or purpose of the step
• Key process parameters (speed, temperature, pressure)
• Inputs and expected outputs

A well-defined process step is the foundation of every subsequent analysis. Vague or overly broad steps lead to incomplete failure identification.

Step 2: Identify Failure Modes

For each process step, identify all potential failure modes — the specific ways in which the step could fail to perform its intended function.

Effective failure modes are specific and measurable:

• Good: "Bore diameter out of tolerance (+0.05 mm)"
• Weak: "Part is defective"

Each process step typically has multiple failure modes. Missing failure modes is one of the most common weaknesses in FMEA practice.

Step 3: Analyze Effects of Failure

Each failure mode must be linked to its effect — the consequence experienced by the customer, end user, or downstream process.

Effects determine the Severity (S) rating, which is the most critical factor in risk evaluation. Effects should describe:

• Impact on product function or safety
• Impact on assembly or downstream operations
• Regulatory or compliance consequences

A failure with no identified effect is not necessarily safe — it may indicate an incomplete analysis.

Step 4: Identify Causes

For each failure mode, identify the root cause or mechanism that leads to the failure. The cause determines the Occurrence (O) rating.

Common cause categories include:

• Machine wear or miscalibration
• Material variation or contamination
• Operator error or insufficient training
• Environmental factors (temperature, humidity)

Jumping directly to actions without identifying root causes is a frequent mistake that reduces the effectiveness of corrective measures.

Step 5: Define Controls (Detection & Prevention)

Controls are the mechanisms in place to either prevent a failure from occurring or detect it before it reaches the customer. The Detection (D) rating reflects how capable current controls are.

AIAG-VDA distinguishes two types:

• Prevention controls — reduce the likelihood of the cause occurring (e.g., poka-yoke, process interlocks)
• Detection controls — identify the failure after it occurs but before delivery (e.g., SPC, CMM inspection, visual checks)

Overestimating detection capability is one of the most common mistakes in FMEA — assuming controls are more effective than they actually are leads to underestimated risk.

Step 6: Define Actions and Priorities

The final step combines Severity, Occurrence, and Detection into a risk evaluation. The AIAG-VDA standard uses Action Priority (AP) instead of the traditional Risk Priority Number (RPN).

Action Priority classifies each failure as:

• High (H) — Immediate action required
• Medium (M) — Action recommended
• Low (L) — Action optional

Unlike RPN (which multiplies S × O × D), AP uses rule-based logic that ensures high-severity risks are never underestimated — even when occurrence is low. For a detailed comparison of Excel-based and structured approaches to managing these priorities, see the FMEA template comparison.

Common Mistakes When Applying FMEA Steps

• Defining process steps too broadly, resulting in missed failure modes
• Using generic failure modes (e.g., "defect") instead of specific, measurable mechanisms
• Skipping root cause analysis and jumping directly to corrective actions
• Treating FMEA as a one-time documentation exercise instead of a living risk management tool
• Using RPN multiplication thresholds instead of AIAG-VDA Action Priority logic
• Assigning overly optimistic Detection ratings without validating control effectiveness

FMEA Steps in Practice

In real-world applications, FMEA steps are typically performed during cross-functional team reviews involving quality, production, and engineering. The process is iterative — as new data becomes available or processes change, the FMEA is updated.

A well-executed FMEA reduces scrap, rework, warranty claims, and audit findings by shifting risk detection upstream — before failures reach the customer.

To see a complete application of these steps, view the FMEA example. To generate a structured FMEA based on your own process, use the FMEA tool.

When Should You Perform FMEA?

FMEA should be performed:

• Before launching a new production process
• When introducing design or process changes
• After major quality issues or customer complaints
• During APQP or process validation phases

In practice, the earlier FMEA is applied, the lower the cost of risk mitigation.

Late-stage detection leads to significantly higher costs due to rework, scrap, or field failures.