Page 114 - Safety Risk Management for Medical Devices

Risk Analysis Techniques 93

Figure 12.11 Information flow between Failure Modes and Effects Analysis (FMEA) levels.

Not every failure has a safety impact. FMEAs can be used for two benefits:
1. for safety risk analysis (Hazard Analysis), and
2. for product risk analysis. Product risks have impact on reliability, perfor-
mance, and project, to name a few, but are distinguished from safety risks.

Each DFMEA/PFMEA at a given level contributes to the next level-up DFMEA.
Failure Modes at a lower level DFMEA/PFMEA become Causes at the next level-up
DFMEA. The End-effects at a lower level DFMEA/PFMEA become the Failure
Modes at the next level-up DFMEA. Probability of Failure Modes at a lower level,
contribute to the probability of a Failure Mode at the next level up (see Fig. 12.11 for
a graphical depiction of this concept).
Per ground rule number 9 in Section 12.4.4, designer errors are excluded from
the DFMEA. Excluding designer errors from the DFMEA does not mean that
designers don’t make mistakes. Designer errors are detected and corrected by process—
which includes peer reviews, modeling, and testing.
It is possible to analyze the design process for ways in which designers could make
mistakes and how those mistakes could escape. But that would be the PFMEA of the
design process. The DFMEA analyzes the output of the design and the ways in which
it could fail.
At the end of the design phase in product development, the DFMEAs should be
transferred to Lifecycle Management and maintained by that department. Risk
management should be kept apprised of any changes to the DFMEAs as part of any
proposed change impact-analysis.

12.6.1 DFMEA workflow
In the following sections the workflow for DFMEA is described. The workflow
corresponds to the template that is provided in Appendix B—Templates.