Premium Resources

Failure Mode And Effect Analysis

An FMEA provides the design engineer, reliability engineer, and others a systematic technique to analyze a system, subsystem, or item for all potential or possible failure modes. This method then places a probability that the failure mode will actually occur and what the effect of this failure is on the rest of the system. If criticality of failure is considered, the technique is called FMECA. The criticality portion of this method allows one to place a value or rating on the criticality of the failure effect on the entire system. It is not uncommon to omit the criticality portion from the methodology.

A FMEA or FMECA (in most cases there is little difference) is a detailed analysis of a system down to the component level. Once all items are classified as to

1) Failure Mode

2) Effect of the Failure

3) Probability that failure will occur, they are rated as per their severity via an index called a RPN (risk priority number)

This RPN is dimensionless from the aspect that there is no real meaning to a value of say 600 versus 450 except in the difference in magnitude. Once all components or items have been analyzed and assigned a RPN value, it is common to work from the highest RPN value down. There will be more discussion on the RPN later.

Failure Mode- The way in which a specific process input fails:

  • Effect - The impact the failure has on the Critical Quality Parameter

  • Cause - The source of variation that caused the process to fail

  • Current controls - Systemized devices in place to prevent or detect the failure

Severity Importance of an effect on critical quality parameter (1-Not severe; 10 very severe)

Occurrence Frequency with which a cause occurs (1-Not likely; 10-Very likely)

Detection Ability of current control to detect the cause before creating a failure mode (1-likely to detect; 10-not likely to detect)

The following are the steps that are taken to prepare an FMEA:

1. FMEA number: This should be a log controlled number assigned for tracking the document

2. The part number, name or other appropriate description

3. The design responsibility: Which department or group is responsible for this design?

4. The person responsible for FMEA preparation

5. The date the FMEA was prepared on and any necessary revision level

6. The subsystem or component part number being analyzed

7. The component function

8. The potential failure mode &the potential effect of failure

10.The potential cause of failure

11. What are the current controls in place to prevent the cause from occurring?

Process Definition Failure Mode and Effect Analysis (FMEA)

  • For ranking 1, the severity has no effect, the occurrence is remote: failure is unlikely less than 1 in fifteen lakhs and detectability has design control that will detect potential cause or the mechanism and subsequent failure mode

  • For ranking 2, the severity has a system operable with minimal interference

  • The occurrence is Low: relatively few failures less than 1 in one lakh fifty thousand. It has very high chance where the design control will detect potential cause/mechanism and subsequent failure mode

  • For ranking 3, the system is operable with some degradation of performance. The occurrence is Low: Relatively few failures less than 1 in fifteen thousand. The detectability has high chance where the design control will detect potential cause/mechanism and subsequent failure mode

  • For ranking 4, the system is operable with significant degradation of performance. The occurrence is moderate: It has occasional failures less than 1 in 2000. It has a low chance of the design control that will detect potential cause/mechanism and subsequent failure mode

  • For ranking 5, the system is inoperable with no damage. The occurrence is moderate with occasional failures of less than 1 in 400. It has a moderate chance where the design control will detect potential cause/mechanism and subsequent failure mode

  • For ranking 6, the system is inoperable with minor damage, the occurrence is moderate with occasional failures with less than 1 in 80 instances. It has low chance where the design control will detect potential cause/mechanism and subsequent failure mode

  • For ranking 7, the system is inoperable with equipment damage. The occurrence is high: repeated failures are less than 1 in 20 instances. It has a very low chance where the design control will detect potential cause/mechanism and subsequent failure mode. Here, there is a very low chance that the design control will detect potential cause/mechanism and subsequent failure mode

  • For ranking 8, the system is inoperable with destructive failure without compromising safety. The occurrence is high: repeated failures are less than 1 in 8 instances. There is a remote chance where the design control will detect potential cause/mechanism and subsequent failure mode

  • For ranking 9, the system is has very high severity ranking when a potential failure mode affects safe system operation with warning. The occurrence is very high where failure is almost inevitable less than 1 in 3 instances. There is a very remote chance that the design control will detect potential cause/mechanism and subsequent failure mode

  • For ranking 10, the system has very high severity ranking when a potential failure mode affects operation without warning. The occurrence is very high: failure is almost inevitable less than 1 in 2 instances. The design control cannot detect potential cause/mechanism and subsequent failure mode