Simulation with both 1 and 2 param Weibull

[Scott]

I am working on sample size estimates of a reliability demonstration test using Monte Carlo simulation (SimuMatic). The life distribution histroically is a Weibull with beta less than 1 and eta in the few million hour range.

Typically for few failures (<5) we force the beta to a historical value but for larger numbers of failures a 2 parameter MLE fit is used. This difference in the analysis isn't possible to specify in the SimuMatic tool, as a result the confidence bounds shown are "wrong". The only way I know to get around this would be to copy and paste all the individual Monte Carlo raw data sets into Weibull++ folios and run the 1-parameter fits on the populations that had few fails.

Anyone know how to get around this?

Thanks,
Scott

[Arai.M]

Hello Scott,
I am assuming since you mention different number of failures, that you are using "Right censoring after a specific time". Unfortunatly, I don't think there is an easy way of doing what you are trying to do in the current Weibull version. I will take this as a suggestion to the development team for future releases.
Thanks,
Arai