Monte Carlo Simulation: how to determine the experimental domain of each factor while keeping a defect rate lower than 5%

Question

Hey JMP users,

I have a question regarding the Monte Carlo simulation.

I need to fix the parameter's experimental domain in order to have a defect rate of 5%.

Is there anyway to proceed by fixing the Defect rate and then deduce how would be the value of mean, lower and upper limit for each factor ?

Thanks !

Byron_JMP · Answer

I could write up a how-to but my colleague @robert_anderson  just did a nice talk on this exact topic at the JMP discovery conference.  I think he nails the solution.

Robust Optimisation of Processes and Products by Using Monte Carlo Simulation Experiments

JMPdiscoverer · Answer

Thanks for your answer.

Yes I need to find the experiemental domain (the largest possible) of my 3 factors while keeping a defect rate less than 5%.

Byron_JMP · Answer

Easy to do as long as you have one mean and one standard deviation, and the mean is constant. As soon as you have multiple standard deviations and the multiple means that aren't fixed, all the parameters are correlated.

It sounds like you're trying to maximize the variation to satisfy the minimum level of acceptance?