cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Choose Language Hide Translation Bar
0 Kudos

Reduce DoE - opposite to Augment DoE

What inspired this wish list request? 

Sometimes when we create DoE and communicate it with experimenters, and discuss its evaluation and go through all the runs/combinations, an SME sees that 1 or more runs could not be done due to some limitations that can not be easily implemented with disallowed filters. Or it can also be that initially there was a capacity to do let's say 16 runs, and we build the DoE which we trust and is the most optimal etc., but due to some last-minute changes in the lab, we have to reduce the DoE to 15 or 14 runs. So we have no time to redo the DoE. 

 

What is the improvement you would like to see? 

My idea would be that there was an option to Reduce DoE, in a way contrary to Augment DoE, so that the algorithm would find the least significant n points and remove it from the DoE table. And recreate all the scripts for modeling and design evaluation.

 

Why is this idea important? 

This would be helpful if we need to make last-minute changes to the experiment strategy.

 

 

4 Comments

If you load in that original design as covariates in Custom Design, you can ask it to pick an optimal subset based on your new run size specifications. You even have the option to force some runs into the final design if they've already been completed. A couple of resources:

 

What is a covariate in design of experiments?

Developer Tutorial - Handling Covariates Effectively when Designing Experiments

aatw
Level III

Thanks

mia_stephens
Staff
Status changed to: Delivered
 
mia_stephens
Staff

Thanks @Ryan_Lekivetz for the suggestion. Marking this as Delivered.