Subscribe Bookmark RSS Feed

PID Tuning with Latin Hypercube Design


Community Member


Mar 16, 2014

Hi everyone.

I am new to JMP and I hope you guys can help. I am working on a deterministic computer simulation experiment and I have collected 40 observations from the simulation, all of which the responses are desirable. So now I would like to study the process and hope to establish a relationship between the variables and the responses. I used the Latin Hypercube Design to generate 30 design points to analyse a process with 3 variables and 2 measurable responses.

In JMP, for the responses, my goal is to minimize the results. The upper limit allowable has been specified as well. I have already set the lower and upper limits for each variable in JMP. Now that JMP has created 30 design points for each variable, when I input those values into the simulation, not all of the results obtained fall within the specified range of desired response. How do I go about this problem? Is there any way I can use the data which I've obtained from my observations as validation data and use it to compare with the data I obtained from the Latin Hypercube Design? All help is deeply appreciated. Thanks.