In some experiments, the levels of factors are difficult to change; then fractional factorial split-plot (FFSP) designs suitable for selection. an FFSP design, divided into two classes—the whole plot (WP) and subplot (SP) factors. selection WP can affect that SP factors, which attracts much attention Paying more a new optimality criterion selecting such is proposed. The robustness proposed meth...

In a recent paper, Monlezun and Blouin (1988) proposed an analysis of covariance for a split plot experiment. They suggest a model consisting of five regression coefficients, one for each of the effects whole-plot treatment means, whole-plot errors, split plot-treatment means, whole-plot and split-plot treatment interactions and the experimental errors. In this paper we critically examine the a...