# zbMATH — the first resource for mathematics

Two-strata rotatability in split-plot central composite designs. (English) Zbl 1341.62254
Summary: A rotatable design [G. E. P. Box and J. S. Hunter, Ann. Math. Stat. 28, 195–241 (1957; Zbl 0080.35901)] for $$k$$ factors is one such that the prediction variance is purely a function of distance from the design center. Of special interest in this paper is the rotatable central composite design (CCD), which most software packages use as the typical default choice for a second-order design. In many cases some factors are hard to change while others are easy to change, which creates a split-plot experiment. This paper establishes that the split-plot structure precludes the possibility of any second-order design being rotatable in the traditional sense. As an alternative this paper proposes the two-strata rotatable split-plot CCD, where the resulting prediction variance is a function of the whole plot (WP) distance and the subplot (SP) distance separately instead of the sum of them. The resulting design is rotatable in the WP space when the SP factors are held fixed, and vice versa. In the special case where the WP variance component is zero, the two-strata rotatable split-plot CCD becomes the standard rotatable CCD.

##### MSC:
 62K15 Factorial statistical designs 62K20 Response surface designs
Full Text:
##### References:
 [1] Box, On the experimental attainment of optimum conditions, Journal of Royal Statistical Society Series B 13 pp 1– (1951) · Zbl 0043.34402 [2] Box, Multi-factor experimental designs for exploring response surfaces, The Annals of Mathematical Statistics 28 pp 195– (1957) · Zbl 0080.35901 [3] Box, Robust designs, Biometrika 62 pp 347– (1975) · Zbl 0312.62061 [4] Box, Factorial designs, the |X’X| criterion, and some related matters, Technometrics 13 pp 731– (1971) · Zbl 0228.62048 [5] Lucas, Optimum composite designs, Technometrics 16 pp 561– (1974) · Zbl 0294.62099 [6] Myers, Response Surface Methodology: Process and Product Optimization using Designed Experiments (2009) · Zbl 1269.62066 [7] Draper, Response surface designs where levels of some factors are difficult to change, Australian and New Zealand Journal of Statistics 40 pp 487– (1998) · Zbl 0923.62083 [8] Letsinger, Response surface methods for bi-randomization structures, Journal of Quality Technology 28 pp 381– (1996) [9] Trinca, Multistratum response surface designs, Technometrics 43 pp 25– (2001) · Zbl 1072.62623 [10] Goos, Optimal split-plot designs, Journal of Quality Technology 33 pp 436– (2001) [11] Goos, D-optimal split-plot designs with given numbers and sizes of whole plots, Technometrics 45 pp 235– (2003) [12] Goos, The Optimal Design of Blocked and Split-plot Experiments (2002) · Zbl 1008.62068 [13] Vining, Response surface designs within a split-plot structure, Journal of Quality Technology 37 pp 115– (2005) [14] Parker, Construction of balanced equivalent estimation second-order split-plot designs, Technometrics 49 pp 56– (2006) [15] Parker, Unbalanced, minimum point equivalent estimation second-order split-plot designs, Journal of Quality Technology 39 pp 376– (2007) [16] Vining, Exact inference for response surface designs within a split-plot structure, Journal of Quality Technology 40 pp 394– (2008) [17] Wang L Vining GG Kowalski SM 2007 [18] Herzberg, Cylindrically rotatable designs of types 1, 2 and 3, The Annals of Mathematical Statistics 38 pp 167– (1967) · Zbl 0171.17004 [19] Huda, Variance functions for m-grouped cylindrically rotatable designs of type 3, Annals of the Institute of Statistical Mathematics 36 pp 481– (1984) · Zbl 0707.62165 [20] Liang, Three-dimensional variance dispersion graphs for split-plot designs, Journal of Computational and Graphical Statistics 15 pp 757– (2006) [21] Kowalski, A modified path of steepest ascent for split-plot experiments, Journal of Quality Technology 37 pp 75– (2005)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.