Mixed Failure Mode Analysis

[Anonymous]

When doing life data analysis for a mixed failure mode sample, which of the two analyses (mixture of two distributions or competing failure modes with censored data) would one prefer and why?

[Tarik]

In the case of competing failure modes, you are aware of the different failure modes that can cause your product to fail (think about it as a deadly disease that compete to kill the human being (males and females), once one of them occur, the person dies). So when entering the data into Weibull++, you will be asked to specify an ID for each failure (to differentiate between the different failure modes). In the case of mixed Weibull think about it as 2 different diseases, Disease A that only affect the male population and Disease B that can only affect female population. When modeling human death, if your population contain both male and females, then you might not see a straight line on the Weibull plot, that would indicates that your population is affected by more than one failure mode at work (ex: Disease that only affect one of genders).


Here are more details from Weibull.com

-Competing Failure Modes:
Often, a group of products will fail due to more than one failure mode. One can take the view that the products could have failed due to any one of the possible failure modes, but since an item cannot fail more than one time, there can only be one failure mode for each failed product. In this view, the failure modes “compete” as to which causes the failure for each particular item. This can be viewed as a series system reliability model, with each failure mode comprising a block of the series system. Competing failure modes analysis segregates the analyses of failure modes and then combines the results to provide an overall model for the product in question.

http://www.weibull.com/LifeDataWeb/competing_failure_modes.htm

- Mixed Weibull:
The mixed Weibull distribution is used to model data that do not fall on a straight line on a Weibull probability plot. Data of this type, particularly if the data points follow an S-shape on the probability plot, may be indicative of more than one failure mode at work in the population of failure times
http://www.weibull.com/LifeDataWeb/the_mixed_weibull_distribution.htm