Failure Mode & Effect Analysis

FMEA is a method used to identify and prioritise potential defects based on their severity, expected frequency and likelihood of detection. The aim is to anticipate problems so that proactive steps can be taken to counteract them and thereby reduce or eliminate the risks. FMEA can be used in many ways:

To improve the design and robustness of a product or process
To improve the design of an improvement solution
To highlights weaknesses in the current design of a product or process
To identify and manage project risk

How to do FMEA

Assemble a cross-functional team of people with diverse knowledge about the target process/product and customers.
Create a FMEA model (e.g. Using ProcessMA > Quality Tools > Templates > FMEA).
For product FMEA, identify the functions (as required by customers) and for process FMEA, identify the process steps.
For each function or process step, identify all the potential ways failure could happen.
For each failure mode, identify all the effects and consequences on the business and its customers.
For each effect, determine how serious it will be when failure does occur. Severity is rated on a scale from 1 to 10, where 1 is insignificant and 10 is catastrophic.
For each failure mode, determine all the potential root causes.
For each cause, determine the how frequently it will occur. Occurrence is rated on a a scale of 1 to 10 where 1 is almost unlikely and 10 is almost every time it will happen.
For each cause, identify existing process controls and mechanism to prevent the cause from happening or reduce its frequency of occurrence.
For each control, determine the ability to detect the cause. This is also rated on a scale from 1 to 10, where 1 means the control is almost certain to detect the problem and 10 means the control has almost no chance of detecting the problem.
Calculate the risk priority number (RPN), which is the product of the ratings in Severity, Occurrence and Ability to detect. The higher the RPN, the greater the risk of the associated failure mode.
Identify actions or changes required to the process or product to lower severity, reduce occurrence or increase ability to detect problems.
When the actions are implemented, update the FMEA on the new ratings for Severity, Occurrence, Ability to Detect and resultant RPN.

Rating Scale

The definition for different ratings of severity, occurrence and ability to detect should be established before the performing FMEA. The definition may vary for different purpose and different organisation. Samples of commonly used rating for service and manufacturing applications are shown as follow:

Rank	Severity	Occurrence	Ability to Detect
10	Safety issue and/or non-compliance with government regulation without warning	Almost every time	Almost impossible
9	Safety issue and/or non-compliance with government regulation with warning	1 in 2	Very remote
8	Operations down for a significant period of time and major financial impact; Loss of primary function	1 in 5	Remote
7	Serious disruption to operations, defects caught at customer site, requires major rework or scrap; Reduction in primary function	1 in 10	Very low
6	Major disruption to operations and requires light rework or scrap; Loss of comfort/convenience function	1 in 100	Low
5	Minor disruption to operations and requires light rework or scrap; Reduction of comfort/convenience function	1 in 500	Moderate
4	Inconvenience to the process and requires minor rework; Returnable appearance and/or noise issue noticed by most customers	1 in 1,000	Moderately high
3	Inconvenience to subsequent tasks and requires minor rework; Non-returnable appearance and/or noise issue noticed by customers	1 in 5,000	High
2	Inconvenience to current task and requires minor rework; Non-returnable appearance and/or noise issue rarely noticed by customers	1 in 50,000	Very High
1	No discernable effect	Almost impossible	Almost certain