Failure Modes and Effects Analysis (FMEA) Procedural Guide


See also:  Failure Modes and Effects Analysis (FMEA) Overview



Describe the product or process.


A clear and specific description of the product or process undergoing FMEA must first be articulated.  The creation of this description ensures that the responsible engineer fully understands the 'form, fit, and function' of the product or process.




Draw a block diagram of the product or process.


A block diagram of the product/process needs to be developed to show the logical relationships between the components of the product or the steps/stages of the process.  A block diagram may be in the form of boxes connected by lines, with each box corresponding to a major component of the product or a major step of the process. The lines correspond to how  the product components or process steps are related to each other.




Complete the header of the FMEA Table (Table 1).


FMEA Table headers vary from one to the next, since they are supposed to be customized according to the requirements of the companies using them. Generally the header requires, among others that you may wish to add, the following information:  Product/Process/System Name, Component/Step Name; Product Designer or Process Engineer, Name of the Person who prepared the FMEA form; FMEA Date; Revision Level (letter or number); and Revision Date.




Enumerate the items (components, functions, steps, etc.) that make up the product or process.


Break down the product or process being subjected to FMEA into its major components or steps. List down each of these components or steps in Column 1 of the FMEA table.  The items must be listed down in a logical manner.



Identify all potential Failure Modes associated with the product or process.


A failure mode is defined as how a system, product, or process is failing.


Now here arises some confusion in the semiconductor industry, which usually measures its failure modes in terms of how the product or process is deviating from its specifications.  A product or process can have hundreds of different failure modes based on this definition, most of which are highly correlated to each other because of a common failure mechanism behind them.  


A failure mechanism is defined as the physical phenomenon behind the failure mode(s) observed, e.g., die cracking, corrosion, electromigration, etc.


To simplify the use of FMEA in the semiconductor industry, therefore, the engineer may choose whether to construct the FMEA table in terms of failure modes or in terms of failure mechanisms.  For convenience of discussion, the term 'failure mode' shall refer to either failure mode or failure mechanism when used in this web page (this web page only!).


An example of a semiconductor process where failure mechanisms may be more effective to use is Wirebonding, whose failure mechanisms include ball lifting, wedge lifting, wire breaking, bond-to-bond shorting, etc.  




List down each Failure Mode using its technical term.


Using an official technical term for listing the failure mode prevents confusion.  All potential failure modes should be listed down for each item (product component or process step). Column 2 of the FMEA Table shall be used for this purpose. 




Describe the effects of each of the failure modes listed and assess the severity of each of these effects.


For each of the failure modes in Column 2,  a corresponding effect (or effects) must be identified and listed in Column 3 of the FMEA Table.   A failure effect is what the customer will experience or perceive once the failure occurs.  A customer may either be internal or external, so effects to both must be included.  Examples of effects include: inoperability or performance degradation of the product or process, injury to the user, damage to equipment, etc.

Assign a severity rating to each effect. Each company may develop its own severity rating system, depending on the nature of its business.  A common industry standard is to use a 1-to-10 scale system, with the '1' corresponding to 'no effect' and the '10' corresponding to maximum severity, such as the occurrence of personal injury or death with no warning or a very costly breakdown of an enormous system. 


Column 4 of the FMEA Table is used for the severity rating (SEV) of the failure mode.



Identify the possible cause(s) of each failure mode.


Aside from its effect(s), the potential cause(s) of every listed failure mode must also be enumerated.  A potential cause should be something that can actually trigger the failure to occur.  Examples of failure causes include: improper equipment set-up, operator error, use of worn-out tools, use of incorrect software revision, contamination, etc. 


The potential causes are listed in Column 5 of the FMEA Table.




Quantify the probability of occurrence (Probability Factor or PF) of each of the failure mode causes. 


The likelihood of each of the potential failure cause occurring must be quantified.  Every failure cause will then be assigned a number (PF) indicating this likelihood or probability of occurrence.  A common industry standard for this is to assign a '1' to a cause that is very unlikely to occur and a '10' to a cause that is frequently encountered.


PF values for each of the failure causes are indicated in Column 6 of the FMEA Table.




Identify all existing controls (Current Controls) that contribute to the prevention of the occurrence of each of these failure mode causes.


Existing controls that prevent the cause of the failure mode from occurring or detect the failure before it reaches the customer must be identified and evaluated for its effectiveness in performing its intended function.  Each of the controls must be listed in Column 7 of the FMEA Table.



Determine the ability of each control in preventing or detecting the failure mode or its cause.


The effectiveness of each of the listed controls must then be assessed in terms of its likelihood of preventing or detecting the occurrence of the failure mode or its failure cause.  As usual, a number must be assigned to indicate the detection effectiveness (DET) of each control.  DET numbers are shown in Column 8 of the FMEA Table.




Calculate the Risk Priority Numbers (RPN).


The Risk Priority Number (RPN) is simply the product the Failure Mode Severity (SEV), Failure Cause Probability (PF), and Control Detection Effectiveness (DET) ratings.  Thus,  RPN = (SEV) x (PF) x (DET).

The RPN, which is listed in
Column 9 of the FMEA Table, is used in prioritizing which items require additional quality planning or action.



Identify action(s) to address potential failure modes that have a high RPN.


A high RPN needs the immediate attention of the engineer since it indicates that the failure mode can result in an enormous negative effect, its failure cause has a high likelihood of occurring, and there are insufficient controls to catch it.  Thus, action items must be defined to address failure modes that have high RPN's.


These actions include but should not be limited to the following: inspection, testing, monitoring, redesign, de-rating,  conduct of preventative maintenance, redundancy, process evaluation/optimization, etc.

Column 10 of the FMEA Tables is used to list down applicable action items.




Implement the defined actions.


Assign a responsible owner and a target date of completion for each of the actions defined. This makes ownership of the actions clear-cut and facilitates tracking of the actions' progress.  The responsible owner and target completion dates must be indicated in Column 11 of the FMEA Table.


The status or outcome of each action item must also be indicated in Column 12 of the FMEA Table. 




Review the results of the actions taken and reassess the RPN's.


After the defined actions have been completed, their over-all effect on the failure mode they're supposed to address must be reassessed.  The engineer must update the SEV, PF, and DET numbers accordingly.  The new RPN must then be recalculated once the new SEV, PF, and DET numbers have been established.  The new RPN  should help the engineer decide if more actions are needed or if the actions are sufficient.


Columns 13, 14, 15, and 16 of the FMEA Table are used to indicate the new SEV, PF, DET, and RPN, respectively.



Keep the FMEA Table updated.


Update the FMEA table every time the product design or process changes or new actions or information cause the SEV, PF, or DET to change.



Table 1. Example of a Simplified FMEA Table




See also:  Failure Modes and Effects Analysis (FMEA) Overview.




Copyright 2005 All Rights Reserved.