TY - JOUR
T1 - A class of three-level designs for definitive screening in the presence of second-order effects
AU - Jones, Bradley
AU - Nachtsheim, Christopher J.
PY - 2011/1
Y1 - 2011/1
N2 - 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 over two-level factors because having three levels allows for some assessment of curvature in the factor-response relationship. Yet, the most familiar screening designs limit each factor to only two levels. We propose a new class of designs that have three levels, 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, and avoid confounding of any pair of second-order effects. Moreover, for designs having six factors or more, our designs allow for the efficient estimation of the full quadratic model in any three factors. In this respect, our designs may render follow-up experiments unnecessary in many situations, thereby increasing the efficiency of the entire experimentation process. We also provide an algorithm for design construction.
AB - 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 over two-level factors because having three levels allows for some assessment of curvature in the factor-response relationship. Yet, the most familiar screening designs limit each factor to only two levels. We propose a new class of designs that have three levels, 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, and avoid confounding of any pair of second-order effects. Moreover, for designs having six factors or more, our designs allow for the efficient estimation of the full quadratic model in any three factors. In this respect, our designs may render follow-up experiments unnecessary in many situations, thereby increasing the efficiency of the entire experimentation process. We also provide an algorithm for design construction.
KW - Alias
KW - Confounding
KW - Coordinate Exchange Algorithm
KW - D-Efficiency
KW - Response Surface Designs
KW - Robust Designs
KW - Screening Designs
UR - http://www.scopus.com/inward/record.url?scp=79251576237&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79251576237&partnerID=8YFLogxK
U2 - 10.1080/00224065.2011.11917841
DO - 10.1080/00224065.2011.11917841
M3 - Article
AN - SCOPUS:79251576237
SN - 0022-4065
VL - 43
SP - 1
EP - 15
JO - Journal of Quality Technology
JF - Journal of Quality Technology
IS - 1
ER -