1
$\begingroup$

I have completed test work following a randomised response surface (multi-factor) experimental design (Central Composite Full) but have found that my levels for two of the factors are not broad enough.

Is there a standardised approach for expanding design of experiments?

I am thinking to: (1) test random central test to validate that I have been able to maintain a consistent standard versus the original experiment tests (using ANOVA / mann-whitney) (2) new tests will simply be those needed to fill the broader Central Composite design (possibly ignoring the ones not needed) (3) test the final model with left over data from the original model to validate

Thoughts?

I am hoping that there might be a standard approach to expanding these kinds of experimental design or perhaps using an alternative design so I can minimise the repeats of experiments.

Improve this question
$\endgroup$
1
  • $\begingroup$ One standar way of expanding an experiment is to use D-optimal designs. The AlgDesign R package (on CRAN) supports that. You should probably have the new runs in an additional block. $\endgroup$
    – kjetil b halvorsen ♦
    Commented Mar 25, 2017 at 17:53

1 Answer 1

Reset to default
0
$\begingroup$

One standard way of expanding an experiment is to use D-optimal designs.

The AlgDesign R package (on CRAN) supports that. You should probably have the new runs in an additional block. Also see the paper “Repairing” Response Surface Designs

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.