Optimal fractional factorial split-plot designs when the whole plot factors are important

In practice, fractional factorial split-plot (FFSP) designs are widely used when the levels of some factors are very difficult or expensive to be changed or controlled. The factors of an FFSP design are divided into two groups, the whole plot (WP) and subplot factors. Under the circumstances that the experimenter has some prior information about the significance of the WP factors, the paper gives a new optimal criterion for selecting designs, that is, the minimum aberration of type WP (WP-MA). Construction of optimal fractional factorial split-plot designs with respect to the WP-MA criterion is considered. The WP-MA FFSP designs with 8, 16, and 32 runs are tabulated.