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