FMEA: Failure Mode and Effect Analysis Description:Failure Modes and Effects Analysis is a tool used by managers toidentify and analyse potential failure risks within the systems. The FMEAprocess begins by identifying “failure modes,” the ways in which a product,service or process could fail.
Team may examine every elementof a service, starting from the inputs till the output is delivered to thecustomer. It is a framework for analysing the design of a product or servicesystem to identify potential failures, then taking steps to counteract or atleast minimize the risks from those failures. These risks are classified accordingto probability of occurrence and impact/ severity, based on experience.
Theobjective of FMEA is to remove ‘identified’ failures out of the system, ideallywith minimum cost, time and effort. It seeks to reduce risk at all levels startingwith most important actions to decrease or limit either impact or probabilityof occurrence of event. (Figure 1)Utility:· Utility of the tool is in ensuringthat any possible disastrous situation is prevented or at the very least, itseffect is minimized and there is a mitigation plan in plan to counter.· Newor existing processes being designed or changed· Carry-over processes for use in newapplications or new environments. Procedureto follow:Processsteps in FMEA:1. Identifypotential failures and effects2. Determineseverity3. Measure likelihoodof occurrence4.
Failure detection5. Generate a Risk PriorityNumberStep 1:Identify potential failures and effectsAnalyse functional requirements and theireffects to identify all failure modes such as fracture. Failure modes in onecomponent can induce problems in others. We need to detect their failure effects.An example could be user injury.
Step 2:Determine severitySeverity is the seriousness of failureconsequences of failure effects. Usually, it is rated on a scale of 1 to 10,with 1 being lowest severity and 10 being hazardous. (Figure 2)Step 3:Measure likelihood of occurrenceExamine how often each failure occurs. Observesimilar procedures and processes and their documented failure modes. Allpotential failure causes should be documented in technical terms. An example ofa failure cause could be misplaced machinery etc.
Again, the ranking is donefrom 1 to 10, with 10 as extremely likely. (Figure 3)Step 4:Failure DetectionAfter mitigation plan is devised, it is testedfor efficacy and efficiency. Inspection is needed, and techniques aredetermined that detect failures. These are assigned a detection value, whichindicate how likely it is to detect failures. Higher values imply that thefailure might not be detected in the system. (Figure 4)Step 5:Generate RPNAll failures are not the same and FMEAprioritizes them according to:Severity: Impact or seriousness of failure.Occurrence: Frequency of default.
Detectability: Level of difficulty in knowing if there is afailure.These failure modes are scored on a scale of 1to 10 and then multiplied to calculate Risk Priority Number (RPN).RPN = Severity X Occurrence X DetectabilityThese results reveal most problematic areas inthe design process.
Higher RPN ratings should be corrected first as they havethe most impact. Managerimplications:1.The idea is to focus our improvement efforts on the failures that are criticaland might have the biggest impact on the customer perceptions. 2.This is a tool to rectify and improve existing design processes in place. Italso helps create a database of failure causes, which can be later referencedto avoid delays. 3.
FMEA can be employed from earliest designto improve and finally, control phase of DMAIC process. References:https://polarion.plm.automation.siemens.com/hubfs/Docs/Guides_and_Manuals/Siemens-PLM-Polarion-How-to-conduct-a-failure-modes-and-effects-analysis-FMEA-wp-60071-A3.pdfhttps://www.isixsigma.com/tools-templates/fmea/minimizing-risks-how-apply-fmea-services/https://www.isixsigma.com/tools-templates/fmea/fmea-preventing-failure-any-harm-done/APPENDIX:Figure1: FMEA example for a car tire problemFigure2: Severity ratings chartFigure3: Occurrence ratings chartFigure4: Detectability ratings chart