FMEA - For Failure Free System

FMEA stands for Failure Mode and Effect Analysis. The premise of this concept is an age old thought ‘Prevention is better than cure’. An FMEA is a systematic study of identifying and preventing problems in a system before they occur. Developed under the aegis of US Military Development programme, of late, has received a lot of ground in other areas such as Space programmes, Automobile sector, Software sector etc. Owing to its wide acceptance and potential in anticipating failures, it was formally inducted into the Quality systems such as QS 9000 and ISO/TS 16949.

Often the failures that are detected late in the system require lot of cost in corrective action. Thus, FMEA can lead to heavy savings in anticipating failures before they occur and prescribing the corrective action. Failure mode is the way in which a system or process or product can fail. FMEA can be done at concept level, design level or for the process of a product or a system. It requires a multidisciplinary team and inputs from all the team members is significant as it helps in correctly anticipating the failures. The anticipated failures can be studied for its effects at three levels viz., Local effect, High Level effect and End effect. It provides flexibility to the team as to how far detail analysis they need to perform. For mission critical systems or complex systems a detail analysis of effects is generally recommended. Effects of a failure can be studied to arrive at the root cause of it. Cause and Effect Analysis is one of the most followed techniques in analyzing the root cause of failures.

One of the most important steps during FMEA is calculating Risk Priority Number (RPN). RPN quantifies the risk carried by each failure. RPN is calculated based on Severity, Occurrence and Detection ranking. Severity is the effect of a failure on the system. Occurrence is the likelihood of a failure. It is the probability that failure will occur this many times in a system. In terms of risk, a failure mode that cannot be detected can present a greater overall risk than those that can be detected. Consequently, in calculating an overall risk priority for a failure mode, it is important to also factor in the chance of detection. All the three ratings are assigned based on the rating scale. RPN is then calculated as a product of Severity, Occurrence and Detection. The RPN number obtained is not a statistically perceived number per se, it provides the level of criticality of those failures based on analysts study of FMEA. It can be treated as historical number to check the criticality of that failure during future analysis.

The aim of FMEA rests on minimizing the risk priority number by recommending solutions for identified failure modes. One of the advantages of FMEA is that it does not require complex statistics per se a brainstorming, rationale and consensus. But, the success of FMEA rests on the experience and expertise of its team and hence can be limitation when the team is naïve. In such scenarios, experience of experts in the field can be applied to do justice with the study.