Screening designs are attractive for assessing the relative impact of a large number of factors on a response of interest. Experimenters often prefer quantitative factors with three levels because it allows for some assessment of curvature in the factor–response relationship. Yet, the most familiar screening designs limit each factor to only two levels.
Introducing....a new class of designs that has three levels. These three-level designs:
- Provide estimates of main effects that are unbiased by any second-order effect.
- Require only one more than twice as many runs as there are factors.
- Avoid confounding of any pair of second-order effects.
This article from The Journal of Quality Technology (reprinted by permission of ASQ) provides an example of algorithm for design structure and recommendations for analysis.