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
Statexplorer
Level III

Mixture design screening

Hi 

 

I have four factors (A, B, C, D); out of that, I need to use only 3 at a time (either A, C, D or B, C, D at a single run) and the sum of three should be 100.

 

How can I do that in a single mixture design

 

 

1 ACCEPTED SOLUTION

Accepted Solutions
Victor_G
Super User

Re: Mixture design screening

Hi @Statexplorer,

 

The following discussion is very similar to your topic and might help you or the other JMP Users create a Custom mixture design respecting the constraint (even if you already found a solution) : Help needed with custom DOE with complex constraints. 

Also with Classical Mixture designs platform there are other options (when/if applicable) :

  • Without factors ranges restrictions and linear constraints: choosing a Simplex Centroïd design with K=3 generates a design with up to 3 factors in each experiment :
    Victor_G_0-1726152814271.png
  • With factors ranges restrictions and/or linear constraints: choosing an Extreme Vertices design of degree K=3 generates a design with up to 3 factors in each experiment :
    Victor_G_0-1726158171667.png

     

Hope this answer may help other users in the Community,

 

Victor GUILLER
L'Oréal Data & Analytics

"It is not unusual for a well-designed experiment to analyze itself" (Box, Hunter and Hunter)

View solution in original post

2 REPLIES 2
Statexplorer
Level III

Re: Mixture design screening

Solved thanks, no need a solution.

Victor_G
Super User

Re: Mixture design screening

Hi @Statexplorer,

 

The following discussion is very similar to your topic and might help you or the other JMP Users create a Custom mixture design respecting the constraint (even if you already found a solution) : Help needed with custom DOE with complex constraints. 

Also with Classical Mixture designs platform there are other options (when/if applicable) :

  • Without factors ranges restrictions and linear constraints: choosing a Simplex Centroïd design with K=3 generates a design with up to 3 factors in each experiment :
    Victor_G_0-1726152814271.png
  • With factors ranges restrictions and/or linear constraints: choosing an Extreme Vertices design of degree K=3 generates a design with up to 3 factors in each experiment :
    Victor_G_0-1726158171667.png

     

Hope this answer may help other users in the Community,

 

Victor GUILLER
L'Oréal Data & Analytics

"It is not unusual for a well-designed experiment to analyze itself" (Box, Hunter and Hunter)