Failure Mode and Effects Analysis (FMEA) is a structured method for identifying potential failure modes, evaluating their effects and causes, considering existing controls, and prioritizing actions. It is one of the risk-management tools identified in ICH Q9.
An FMEA reviews a process, function, or component systematically. Many methodologies assign ratings for severity, occurrence, and detection, although the scales and scoring rules should be defined by the organization.
Some FMEA methodologies combine the ratings into a Risk Priority Number (RPN) or another prioritization measure. The resulting score can support prioritization, but it should not replace consideration of individual factors, particularly severity, or the judgment of knowledgeable subject-matter experts.
FMEA can provide useful structure for complex processes, novel systems, or unclear failure paths. It is not mandatory for computerized-system validation. A simpler risk-assessment method may be appropriate when it provides sufficient information to support the assurance decision.
Frequently Asked Questions
How is the RPN calculated?
In FMEA methodologies that use an RPN, it is commonly calculated by multiplying severity, occurrence, and detection ratings. The scales are defined by the organization, so the possible range depends on the chosen scoring model. RPN should support, not replace, expert evaluation.
Is FMEA required for computer system validation?
No. FMEA is one available risk-assessment method. GAMP 5 supports an appropriate and proportionate risk-based approach rather than mandating a particular tool.
What is the difference between FMEA and a risk assessment?
Risk assessment is the broader activity of identifying, analyzing, and evaluating risk. FMEA is one structured method for performing that activity by examining potential failure modes and their effects.
When is FMEA the wrong tool?
FMEA may be unnecessarily formal when the system is simple, uncertainty is low, failure paths are well understood, and a simpler method provides sufficient information for the decision. The tool should be proportionate to the question and risk.