by L Hurley

Created on: September 26, 2009

Failure mode and effects analysis, (FMEA), is used to anticipate and prevent failures in the products or services produced by a business. By preventing failures a business can improve the performance, quality, reliability and safety of its products which results in improved reputation and increased sales.

How Does FMEA Work?

FMEA uses a number of terms and acronyms that can make it seem complex for the first time user. It is however a straightforward undertaking that only requires a good knowledge of your business and simple arithmetic skills.

FMEA assesses the steps of a process to identify where problems may occur. The problems are scored using a rating scale to determine which problems pose the biggest risk to the business. A high score indicates the need to find and put in place preventative measures to avoid problems.

What Does FMEA Involve?

FMEA uses a combination of brainstorming and numerical scoring through a series of steps to assess the severity, likelihood of occurrence and detectability of problems. The assessment results in a Risk Priority Number, (RPN), that indicates the urgency of the need to address the problem.

FMEA Steps

The first step in undertaking FMEA is to list all the process steps in producing a product. For each process step you then consider what could go wrong. These possible problems are called failure modes and each process step could have more than one failure mode.

The next step is to identify the consequences of each failure mode should it occur. On a scale of one to ten with ten being the worst you score the severity of the consequence.

For example if a failure mode could result in injury to a staff member or customer you would probably rate it at ten. If however the consequence was that the packaging was a slightly lighter shade of blue than it should be you may only score it at one or two.

Next you need to identify the causes of the failure modes and score the likelihood of them happening. If the failure is likely to happen more than once a day you may score it at ten. A failure once a year may only merit a score of one.

The last score is based on how detectable the failure will be. If it is obvious and the product can be pulled before reaching the customer then it would score one. If however the failure could not be detected until the customer used the product it would receive a score of ten.

Once the three scores for each failure mode have been determined they are multiplied to obtain the RPN. Severity times occurrence

• 1
• 2
• Next Page >>