cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Browse apps to extend the software in the new JMP Marketplace
Choose Language Hide Translation Bar

JMP Two level Screening is giving me a different Design every time

Hi! I am very new to JMP just as an fyi.. But I am trying to utilize the DOE Two-level Screening function. I had already gotten a design using this function however,  I am now trying to recreate this same output. I used the same factors and the same amount of runs yet I am getting a different outcome. Is this just a part of how the JMP software works?

1 ACCEPTED SOLUTION

Accepted Solutions

Re: JMP Two level Screening is giving me a different Design every time

After you enter your factors, you need to answer a question:

Dan_Obermiller_0-1709000290595.png

I think you are choosing to construct a main effects screening design. That is an approach that will build an orthogonal or near-orthogonal design. As such, you very well could get different results each time, especially if an orthogonal design does not exist given your situation. If you select to Choose from a list of fractional factorial designs, you will get the more traditional full and factorial designs which will be the same each time.

 

I would encourage you to read about these screening designs in the JMP help: 

https://www.jmp.com/support/help/en/17.2/?os=win&source=application#page/jmp/overview-of-screening-d... 

 

 

Dan Obermiller

View solution in original post

3 REPLIES 3
statman
Super User

Re: JMP Two level Screening is giving me a different Design every time

Welcome to the community.  The default when you are creating a design is to randomize the run order.  If you choose a different selection (like keep the same) you will get the same design. 

Screenshot 2024-02-26 at 12.36.41 PM.jpg

"All models are wrong, some are useful" G.E.P. Box

Re: JMP Two level Screening is giving me a different Design every time

Hi! Thanks for your help!

 

I just tried and I do not have the same options as the image you posted, I have attached a screenshot, so I inputted the same amount of runs as the previous design I am trying to replicate. On the next screen after I select Make Design is where I can choose Keep the same, however the design has already been made so all it seems to be changing on my end is the order in which it shows up on the design table. But the design as a whole is completely different. Even the color matrix on correlation is completely different. Screenshot 2024-02-26 144716.pngScreenshot 2024-02-26 144748.png

Re: JMP Two level Screening is giving me a different Design every time

After you enter your factors, you need to answer a question:

Dan_Obermiller_0-1709000290595.png

I think you are choosing to construct a main effects screening design. That is an approach that will build an orthogonal or near-orthogonal design. As such, you very well could get different results each time, especially if an orthogonal design does not exist given your situation. If you select to Choose from a list of fractional factorial designs, you will get the more traditional full and factorial designs which will be the same each time.

 

I would encourage you to read about these screening designs in the JMP help: 

https://www.jmp.com/support/help/en/17.2/?os=win&source=application#page/jmp/overview-of-screening-d... 

 

 

Dan Obermiller