used to test variances or dispersion effects. Fpr example, if a company has two manufacturing lines making the same product but different regions of the world, a F-test may be used to compare the variance of one line's output vs. that of the other. [Statistical inference tests to compare the quality of different products / processes and compare the performance of different groups.]
a statistical method used to describe variability among observed variables in terms of fewer unobserved variables.
Factorial design are generally employed in engineering and manufacturing experiments. It is appropriate when several factors are to be investigated at two or more levels and interaction of factors may be important. Also see Design of Experiments.
Full Factorial Designs – study of two or more factors where all possible combinations of the levels of the factors are investigated. In brief, 2k Full Factorial involves: a) an experiment with k factors; b) each factor is studied at 2 levels (high-low); c) investigate all combination of factors and levels. Note: 3k full factorials studies k factors, each at 3 levels (high, medium, low).
an event when a process or machinery is not able to perform scheduled operations to specification. Failures are two types - catastrophic (random - without warning) failures and wear-out (slow degradation) failures. Failures can be attributed to operating a process outside its performance parameters, product / process variability and lack of proper maintenance.
is a systematic examination of a failed process, equipment or product. Tools used in failure analysis - failure mode analysis, fault tree analysis, failure mode and effects analysis, 8D and 5-Why process.
Failure Mode Analysis - FMA
Developed from a fault tree, a FMA chart is usually prepared to assess the probability of and assign priority to potential root causes of failure. FMA helps analyze the ways a component or system has failed.
Failure Mode Avoidance
a quality discipline with focus on finding failure modes early in the development process, then applying countermeasures to fix them. D Clausing suggested FMA as a pragmatic strategy to achieve reliability improvement.
Failure Mode and Effects Analysis - FMEA
Failure Mode and Effects Analysis is systematized technique which identifies and ranks the potential failure modes of a design or manufacturing process in order to prioritize improvement actions. FMEA useful in measure & improve phase of DMAIC and characterize in DFSS. Provides information in quantifying weighted priorities, risks and contributing characteristics of a problem. Helps drive team consensus. Click here for a small presentation from WISC | FMEA form | developing a FMEA | SAE J-1739 Recommended Practice for FMEA
Fault Tree Analysis
A deductive analysis method that provides a systematic description of the combinations of possible occurrences in a system that can result in failure. It is a graphical representation of the boolean logic that relates to the result. The method can be applied as a diagnosis and development tool and is especially helpful in early design stages. This was developed in 1962 by HA Watson (Bell Laboratories) for the US Air Force. Boeing adopted this in 1966.