Fail-Safe FMEA

Abstract:Failure mode and effects analysis (FMEA) is a useful risk analysis tool, but is not sufficient by itself to manage risk. The U.S. Food and Drug Administration has issued guidelines on the use of FMEAs. An effective FMEA consists of three tools: a process map, a cause and effects matrix, and the FMEA proper. This FMEA identifies preventive actions an organization can take to eliminate risks. When failures do occur, a corrective action and preventive action (CAPA) system can be informed by the FMEA to allow the organization to quickly respond to the failure. The CAPA should then inform the FMEA, further improving the overall risk management system. An FMEA which is not integrated with performance measures or lacks definitions for severity, occurrence, or detectability, can perpetuate quality …

Access this article
Other ways to access this article

Social Bookmarking

Digg, delicious, NewsVine, Furl, Google, StumbleUpon, BlogMarks, Facebook



good article
--salim alrawahi, 01-27-2012


As a member of the committee that developed the medical device risk management standard ISO 14971 and adjunct faculty at Virginia Tech's Health Products Risk Management Graduate Program, I find that I often have to caution those using FMEA to remember that it was designed as a reliability tool. It is NOT risk analysis, however it is a good tool to use after the product is designed to find if all of the hazards in the fault condition are identified as required in ISO 14971. The pharma guidance, ICH Q9 which was derived from ISO 14971, also uses FMEA as one tool to aid in risk management.

It is important that we identify the definition of risk as it is used in a variety of contexts. In both ISO 14971 and ICH Q9, direction is provided for product safety and is not intended to address business or compliance or other types of risk. For the latter and additional types of risks, ISO 31000 provides definitions and direction.

As FMEA identifies the hazard and the effect of a hazard, it operates on a different level from risk analysis, which analyzes the severity of harm and probability of occurrence of harm. This provides a different result from FMEA.

The standard also requires identifying all of the normal condition hazards, which FMEA does not do. Normal condition hazards are usually identified in the usability engineering activities.

Regulators require that early in the design process, at Design Input, safety requirements (those from risk analysis) are identified. Since FMEA requires some design be complete, it is not usual to have inputs from this tool at the Design Input stage. As inputs are revised at later points, the inputs from FMEA are applied.

If a good job is done in risk analysis at the beginning of design, many of the risks may be identified and input into the design stage, reducing the impact of costly late stage design changes.

So while FMEA is a valuable tool when used properly, it can reduce the effectiveness of risk management when used improperly.

This article provides good information on use of FMEA, but the cautions I have attempted to provide may give good direction to the reader in using the tool for good product risk management.
--Edwin Bills, ASQ Fellow, 01-18-2012


Not using a C&E matrix (simplified extraction from Quality Function Deployment) makes the FMEA a long, drawn-out process that demoralizes participants. Any Six Sigma professional knows this.
--Chris Idzerda, 01-11-2012


I too liked the incorporation of the cause & effects matrix into the mix as a preliminary step in creation of an FMEA. As noted, I've usually gone directly to the FMEA from a process map.
--Larry Norris, 01-09-2012


Fairly good article for as far as it went. Unfortunately the authors focused on only 1 tool (process map) for identifying the steps in a process and 1 tool (C&E matrix) on prioritizing. It could be improved by offering more tools to be used at various steps.
--Bruce Bader, 01-09-2012


Good article. I was particularly interested in the explanation of the cause and effect matrix. Our company calls it a correlation matrix and just counts the number of boxes that have a number in them. That never really made sense to me. The cross-multiplying and addition makes much more sense. It shows where the critical areas are that you should focus on first.
--Tim Imhoff, 01-05-2012


Featured advertisers