Multifaceted Evaluation of 2-Level Supersaturated Designs
2016 Volume 2 Issue 1 Pages 48-59
Details
2016 Volume 2 Issue 1 Pages 48-59
Experimental design is a statistical technique used in the field of quality control. It is applied to determine the relationships between intended results and the factors considered to influence them. It is a fractional factorial design that is suitable when there are many factors considered to influence the result.
Recently, many supersaturated designs have been studied. These are experimental designs, which can be assigned more factors than the number of experiments. The technique is applied to reduce the number of influencing factors from the total set.
Previous studies focused on structure designs and their respective analyses. Each constitution method is evaluated absolutely particularly. Consequently, prior to this study, the proposed designs had not been systematically evaluated. This paper examines some previously proposed supersaturated designs with the use of several indices. It presents a selection guide that describes recommended constitution methods.
