University of Minnesota
Definitive Screening Designs (DSDs), discovered in 2011, are a new alternative to standard two-level screening designs. There are many desirable features of this family of designs. They require few runs while providing orthogonal main effects and avoiding any confounding of main effects by two-factor interactions. In addition they allow for estimating any quadratic effect of the continuous factors. The two-factor interactions are correlated but not confounded with each other. Moreover, in DSDs with 6 or more factors, it is possible to fit a full quadratic model in any three factors. So, if three or fewer factors turn out to be important, these designs can morph from screening to response surface designs allowing optimization without requiring more runs.
This talk will show how to create these designs and adapt them for the presence of two-level categorical factors. It will also demonstrate how to block DSDs orthogonally.
Analysis of DSDs is complicated somewhat by the correlation of two-factor interactions with each other and with quadratic effects. The talk concludes with some new ideas for obtaining clear cut analytical results.
More information about Christopher Nachtsheim may be found at http://www.carlsonschool.umn.edu/faculty-research/nacht001/Christopher_J_Nachtsheim.aspx