Failure Mode Effects Analysis (FMEA)...

Information | Understanding | Best Practice.

Failure Mode Effects Analysis (FMEA) is used to identify the risks which may arise with a product or process and the resultant effects which may be experienced from those risks. FMEA is a bottom-up approach to failure analysis, for example you consider a failure in a component of a product and assess the impact of the failing component on the final product which contains the component. Equally consider the failure of some individual aspect within a process, what will be the resultant impact on the overall process itself.

FMEA - Failure Mode Effects Analysis.

Continuous improvement utilizing Analytical Techniques.
Information | Understanding | Best Practice   >>>

During the FMEA analysis, inputs such as the components utilized, or process steps are analysed and from this analysis comes an output in the form of a risk analysis table from which can be developed a risk management plan. The advantage of the FMEA process is that it is an in-depth process, which forces a detailed analysis of the potential risk points. It is also widely used, accepted and understood. A key disadvantage is that it is restricted to a single failure mode analysis, i.e. in reality risks and failure events often arise due to a combination of reasons, the FMEA only considers single failure conditions, therefore needs to be used as part of an overall risk management process.


How to perform a Failure Mode Effects Analysis (FMEA)?

The first step is to clearly define the scope of the failure analysis, detailing the internal and external interfaces with the process (or product) to be assessed and the internal processes themselves. A clearly outlined and understood scope is critical to the success of the FMEA to be performed.

Then flowchart the process, detailing all the various inputs and outputs from the process. You need to detail all interfaces with other processes or products.

For all stages within the process identified you need to identify all the potential individual failure modes and all the potential interface failure modes.

For each potential failure mode identified, an impact assessment needs to be determined. The impact will be considered in terms of (say) safety, quality, reliability, financial, reputation, impact on the customer or on the business itself, the community, or any other stakeholder deemed important.

Finally, for all the failure modes identified, document the potential severity and expected probability of occurrence. In some organizations, the assessment will also consider the detectability of the failure event. This is not essential, but can add to the value of the FMEA process.


Versions of Failure Mode Effects Analysis. 

All are based on similar principles and a similar approach, however, with differing areas of focus.


A process FMEA focuses on production processes, manufacturing processes or assembly processes.


The component FMEA, looks at the components utilized in a product and the risks associated with component failure or non-conformance.


The design FMEA focuses on product design, where the risks associated with the design of the product or process are assessed.


An application FMEA, looks at the risks associated with the use or application of a product.

Failure Mode Effects & Criticality Analysis (FMECA)
An FMECA is used to identify the potential failures in terms of their criticality. The greater the potential impact, then the more critical it becomes to minimize or prevent the potential risks from arising.


Allocating Risk levels based on Failure Mode Analysis.

For all failure modes identified a risk prioritization number (RPN) can be determined. The number is based on the expected “severity” of the risk if it were to arise, the “probability” of the risk arising and the “detectability” of the risk.

The RPN number is determined by the formula:

RPN = S x P x D


RPN = Risk Prioritization Number
S = Severity
P = Probability
D = Detectability

Severity is the impact on the user, the organization, etc.. As outlined previous, this can be associated with a process safety issue which potentially impacts an employee, a product quality issue which impacts a user, a financial or reputational impact on the organization, environmental impact on the community, etc.

The Probability relates to the likelihood that the failure mode could arise. This may be calculated as (say) 1 in 1000, or once every 5 years, etc..

The Detectability relates to the ability of the organization to become aware of the failure having arisen. For example, a machine fault on a production line may be immediately detected, therefore while the severity and probability may be high, the detection will mitigate the overall risk to the organization. Alternatively, contamination on a food processing line, may not be detected until the product has been used by the customer. Here the potential risk to the organization associated with the later detection of the process failure, is clearly greater.

The greater the severity and the higher the probability, the greater the risk to an organization. The lower the detectability, then the greater the risk to the organization.

By calculating the RPN number, the areas of focus are clearly identified. Also, the RPN number can be recalculated after process or product design changes, to determine if the overall risks have been reduced or increased. From an organizational perspective, management can take a policy position that no risk would have an RPN above a certain pre-determined level. This then can then feed into a corporate strategy for risk reduction.

TQM Tools and Techniques Full Details

Quality Improvement Techniques

Quality Management Tools and Techniques …

        • Continuous improvement utilizing Analytical Techniques.
        • Brainstorming
        • 5 why’s analysis
        • Process Flow Diagrams/Flowcharts/Process Mapping
        • Check sheets /Check Lists
        • Run charts
        • Histograms
        • Scatter Diagrams/Scatter Plot
        • Cause and Effect/Fishbone/Ishikawa Diagrams
        • Identifying sources & causes of variation
        • Control/Shewart Charts/DPU Charts
        • Cpk and Ppk Analysis
        • Pareto Analysis
        • Bottleneck Analysis
        • Benchmarking
        • FMEA
        • FTA
        • HAZOP
        • SIPOC
        • Etc. Etc.
        • Information | Understanding | Best Practice.   >>>