- slide 1 of 2
FMEA is an acronym for Failure Modes and Effects Analysis, a tool used to predict risks. Using the FMEA method, an RPN (Risk Priority Number) is calculated by multiplying the three categories of Severity, Occurrence and Detection. Each one assigned a value between 1 and 10, where 10 is high.
While this is a simple method and can allow teams to prioritize in order of highest RPN number, it doesn't necessarily mean that it's an effective method that takes today's high standards into account. FMEA, like any process is prone to failure if its weaknesses are not recognized and supplemented with other activities, such as strategic error-proofing (SET), as is the case in the automotive industry.
The greatest limitations of FMEA are:
- A Bias Toward Severity Ratings. To take an extreme example; Manufacturer Smith makes industrial power supplies and, as part of the FMEA process, a potential failure mode "reversed voltage regulator VR1" is identified. The severity is rated 10, the highest rating possible because the component overheats and could explode, causing harm to the customer. The occurrence is low - rated 1 - as it very rarely happens. The detection is also rated a low 1 since it is easily detected by the operator who has to manually add a heatsink to the device. No improvement is possible on this item as the severity rating will always remain 10. This is a common problem with FMEA since Detection and Occurrence can always be reduced but it is only in rare cases that Severity ratings can be reduced. In this example , the failure mode will always have a 10 Severity rating.
- Problem of Scope. Once the severe issues are remedied, to improve further, where is the cutoff point for sub-processes and components? Obviously it is not feasible for most companies to perform a "what if?" analysis on a circuit board with thousands of discrete components – if this resistor is missing, then that will happen, and so on. Faults such as those are left to debug technicians, and a failure database is collected outlining issues. Severe problems should be added to the FMEA control sheet but trivial issues have no place on it. If every issues ends up on an FMEA sheet then FMEA meetings become very long, tying up hours during which actual processes could be improved.
- slide 2 of 2
- An Ever-Increasing List of Possible Failure Modes. The list of failure modes will grow exponentially and is only limited by the ingenuity and experience of your team members. This means of course that FMEA is simply a list of possible failures, "what can go wrong?" and not a process that forces the correct processes to be implemented.
- FMEA Is Only As Good As the Members of the FMEA Team. Issues that are unexperienced by team members will not be presented as possible failure modes. When these occur, reactive investigation of the problem may even require the hire of external consultants.
- If FMEA Is Not Started at the Design Stage, Its Effectiveness Is Weakened. This is self explanatory.
- Experienced Operators Should Be Part of the FMEA Team as their input can provide valuable insight to process errors.
- Detailed Analysis Is Necessary For FMEA. An overview of a process is not enough.
- If Detection Controls Are Not Tested Adequately (by skipping stages in the process for example) then your detection system may not be adequate. FMEA will not identify this problem.
- FMEA Is an Assessment Tool, it doesn't eliminate problems, and it should be used with error-proofing tools to have the desired effect.
The most important thing to remember, in order to reduce the limitations of FMEA, is to always update the FMEA sheet and related control plans. Failure to do so can have serious consequences, when a severe problem is not documented or ends up causing harm to a customer. FMEA is an important part of any production process, but needs to be supplemented with other project management methodologies to be completely effective.