cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Check out the JMP® Marketplace featured Capability Explorer add-in
Choose Language Hide Translation Bar

Replicate runs in DoE model

Hi, 

I have created a custom DoE design in JMP and specified to have 3 replicate runs in the section Design Generation. However, after making the design I can only find 2 identical runs. Why so?

5 REPLIES 5
statman
Super User

Re: Replicate runs in DoE model

If you examine your design, you should have 3 different treatment combinations that have replicates (2 identical treatment combinations for 3 treatments).  Adding replicates does not take one treatment and replicate it 3 times.

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

Re: Replicate runs in DoE model

Hi @LargeWordsPuma5,

In addition to the response from @statman, there is a slight difference in the terms used by JMP depending on the DoE platform used that explain the different results (and may create confusion):
- In a "Classical design", JMP displays "Number of Replicates", meaning here the number times to perform each run (in addition to the original runs),
- In a "Custom design" JMP displays "Number of Replicate Runs", meaning here the number of run(s) to repeat (so you'll have two identical treatments combination for the number of "replicate runs" specified).

I hope this will help clarify,
Victor GUILLER
L'Oréal Data & Analytics

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

Re: Replicate runs in DoE model

Hi Victor, 

Thank you for your answer. 

Due to the potential biological variation in my experiment (and requirements of the journals) I want to make three technical replicates of each treatments. Is there a way to include those in the design itself using custom design? 

 

Thanks in advance! 

 

 

statman
Super User

Re: Replicate runs in DoE model

Much depends on what questions you want to be able to answer regarding the variation (biological and other noise).  If you just run each treatment in the initial design 3 times randomly, you will be able to analyze that data in JMP and the variation captured by the replicates will be used to estimate the random errors for statistical tests (e.g., mean square error in ANOVA).

 

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

Re: Replicate runs in DoE model

Hi @LargeWordsPuma5,

One option could be to create your Custom Design without any replicates, and then use the platform "Augment Design" to create the replicates of your initial design : https://www.jmp.com/support/help/en/17.1/index.shtml#page/jmp/replicate-a-design.shtml#ww76833

I hope this answer will help you,

Victor GUILLER
L'Oréal Data & Analytics

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