Answers are grounded in this paper and cite their sources.
Mixed two- and three-level designs are widely used in both computer experiments and physical experiments. According to the effect hierarchy principle of factors, the lower-order effect is more important than the higher-order case, especially the main effect of factors. This paper primarily focuses on the two-dimensional projection uniformity of mixed two- and three-level designs. Firstly, a measure of two-dimensional projection uniformity is defined for asymmetric designs. Secondly, the analytical connection is established between this measure and the centralized L1-distance for mixed two- and three-level designs, and a tight lower bound of the measure is derived, serving as a benchmark for evaluating two-dimensional projection uniformity of designs and searching for uniform two-dimensional projection designs. Thirdly, the analytical connection is established between the correlation coefficient and the L2-distance of mixed two- and three-level designs is provided. Finally, the application of a uniform two-dimensional projection design in OTC push-pull circuits is provided for explaining its excellent performance.
No takes yet. Share an insight, caveat, or question.
Xie et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: