3.3 Failure Mode and Effects Analysis

3.3 Failure Mode and Effects Analysis (FMEA)

Failure Mode and Effects Analysis (FMEA) is a proactive, structured methodology used by quality engineers to identify potential failure modes in a design or process, evaluate the risk associated with those failure modes, and rank order recommended corrective actions. First developed by the U.S. Military in the late 1940s (MIL-P-1629) and later popularized by NASA and the automotive industry (QS-9000 / AIAG), FMEA is a core tool in Advanced Product Quality Planning (APQP).

---

1. DFMEA vs. PFMEA Comparison

FMEA is primarily applied in two distinct domains during product deployment: Design FMEA (DFMEA) and Process FMEA (PFMEA).

+-----------------------------------------------------------------------------------+
|                                 FMEA TYPES                                        |
+-----------------------------------------+-----------------------------------------+
|            DESIGN FMEA (DFMEA)          |           PROCESS FMEA (PFMEA)          |
+-----------------------------------------+-----------------------------------------+
| Scope: Component & system functions     | Scope: Manufacturing & assembly steps   |
| Focus: Design deficiencies, geometry,   | Focus: Human error, machine wear,       |
| material selection, stress thresholds   | tooling misadjustment, environment      |
| Assumes: Manufacturing is executed      | Assumes: Product design is correct      |
| according to specifications             | and meets functional intent             |
+-----------------------------------------+-----------------------------------------+

Comprehensive Comparison Matrix

---

2. FMEA Risk Evaluation Metrics (Severity, Occurrence, Detection)

Three rating scales, each scored on a 1-to-10 scale, are used to evaluate risk:

1. Severity (S): Assessment of the seriousness of the effect of a potential failure mode on the customer or downstream user ($1 = \text{negligible impact}$, $10 = \text{hazardous failure without warning}$).
   Note: Severity relates ONLY to the effect of the failure mode.
2. Occurrence (O): Assessment of the likelihood that a specific cause will occur during the product's design life or process execution ($1 = \text{extremely unlikely / } < 1 \text{ in } 1,500,000$, $10 = \text{almost certain / } > 1 \text{ in } 10$).
3. Detection (D): Assessment of the capability of current design or process controls to detect the failure cause or failure mode BEFORE the product leaves the design organization or manufacturing facility ($1 = \text{certain detection / automated poka-yoke}$, $10 = \text{absolute uncertainty of detection / no control}$).
   Note: Lower Detection numerical values represent BETTER detection capabilities.

---

3. Risk Priority Number (RPN) & AIAG & VDA Action Priority (AP)

Traditional RPN Calculation

The Risk Priority Number (RPN) is the mathematical product of Severity, Occurrence, and Detection:

$RPN = S \times O \times D$

Because each factor ranges from 1 to 10, the RPN ranges from 1 to 1,000.

Worked Numerical Example

Consider an automotive brake caliper design evaluated in a DFMEA:

• Failure Mode: Hydraulic fluid seal degradation at low ambient temperature ($-40^\circ\text{C}$).
• Failure Effect: Loss of braking capability leading to vehicle collision ($S = 10$).
• Failure Cause: Seal polymer glass transition point exceeded ($O = 4$).
• Current Detection Control: Standard room-temperature pressure test ($D = 6$).

Initial RPN Calculation:

$RPN_{\text{initial}} = 10 \times 4 \times 6 = 240$

Recommended Mitigation Action:

Redesign seal using a fluorosilicone elastomer (lowers glass transition temperature) and implement low-temperature qualification testing.

Revised Scoring Post-Mitigation:

• Severity ($S$): Remains 10 (The severity of losing brakes does NOT change unless the system architecture changes).
• Occurrence ($O$): Reduced to 1 (Material change eliminates cold-weather degradation).
• Detection ($D$): Reduced to 2 (Low-temperature qualification test guarantees detection).

Revised RPN Calculation:

$RPN_{\text{revised}} = 10 \times 1 \times 2 = 20$

Net RPN Reduction: $240 - 20 = 220$ point risk reduction ($91.7%$ risk reduction).

---

AIAG & VDA Action Priority (AP) Framework

Traditional reliance on RPN thresholds (e.g., "take action if $RPN > 100$") had a fatal flaw: an item with $S=10, O=2, D=4$ ($RPN = 80$) would be ignored, while an item with $S=4, O=5, D=6$ ($RPN = 120$) would trigger corrective action, despite the former carrying catastrophic safety risk.

The unified AIAG & VDA FMEA Handbook (2019) introduced Action Priority (AP) tables, which prioritize risk based on logic trees prioritizing Severity first, then Occurrence, then Detection:

• High (H) Priority: High severity ($S = 9-10$) paired with moderate-to-high occurrence or detection. Mandatory action required.
• Medium (M) Priority: Moderate severity or lower occurrence/detection. Action recommended to lower risk.
• Low (L) Priority: Low risk across all dimensions. Optional action.

---

4. CQE Exam Tips & Key Pitfalls

[!IMPORTANT] A mandatory CQE exam rule: Severity (S) can ONLY be reduced by changing the physical design geometry, material, or system function. Adding an inspection, testing protocol, or SPC chart does NOT change Severity—it only lowers the Detection (D) or Occurrence (O) score!