Share your ideas for the JMP Scripting Unsession at Discovery Summit by September 17th. We hope to see you there!
Choose Language Hide Translation Bar
Level I

Hard to change factors, center points and replicates

I am making a design where I want to use the "hard to change" option for three of my four factors for practical reasons, and I want to add a few center points (which will be from historical data) to the design and make replicates of the design points. Is there some way to implement this in JMP while using the "hard" option?


Re: Hard to change factors, center points and replicates

This design should not be difficult, as far as you have described it, using DOE > Custom Design.


I would caution you about adding replicates. Replication automatically appears when there are sufficient number of runs above the minimum. You can, though, request a minimum number of replicates, too.


I would also caution about using historical data for the center points. If you just mean the factor levels, that should be no problem. If these levels do not appear in the optimal design, then use discrete numeric factors instead of continuous factors. If you mean to enter historical responses, then you run the risk that something is different during the experiment but you won't know. Besides estimation the center points can also serve as control runs.but only if they are run again in the experiment at the same time as the other runs.


See Help > Books > Design of Experiments. The explanation of custom design in the early part of the guide should make it clear how to handle each of your requests.

Learn it once, use it forever!
Article Labels

    There are no labels assigned to this post.