Optimal Blocking Of Two-Level Fractional Factorial Designs With General Minimum Lower Order Confounding

The inhomogeneity of the experimental units will increase the variation of the experiment and lower the efficiency of the experiment. When experimental units are inhomogeneous, the best way is to block the experimental units into different groups. Two situations are taken into consideration. One considers a single block variable and the other considers multi block variables. In view of the two situations, Wei, Li and Zhang (2014) and Zhang, Li and Wei (2011) respectively proposed B1-GMC criterion and B2-GMC criterion. Zhao, Li, Zhang and Karunamuni (2013) proposed a construction theory and obtained the B1-GMC designs. So far, many researchers studied the issues of a single block variable. However, few researchers considered the issues of multi block variables.This paper considers the optimal blocking of fractional factorial designs with gen-eral minimum lower order confounding.Chapter1mainly introduces a single block variable, multi block variables and the difference between a single block variable and multi block variables.Chapter2mainly introduces the optimal blocking of two-level fractional factorial designs with general minimum lower order confounding in the case of three two-level block factors.Chapter3includes some examples to illustrate the optimal blocking of two-level fractional factorial designs with general minimum lower order confounding in the case of three two-level block factors.Chapter4gives a brief concluding to the whole paper.