Optimal Fractional Factorial Plans for Asymmetric Factorials
نویسندگان
چکیده
Fractional factorial plans for asymmetric factorial experiments are obtained. These are shown to be universally optimal within the class of all plans involving the same number of runs under a model that includes the mean, all main effects and a specified set of two-factor interactions. Finite projective geometry is used to obtain such plans for experiments wherein the number of levels of each of the factors as also the number of runs is a power of m, a prime or a prime power. Methods of construction of optimal plans under the same model are also discussed for the case where the number of levels as well as the number of runs are not necessarily powers of a prime number. 1. Introduction. The study of optimal fractional factorial plans has received considerable attention in the recent past; see e.g., Dey and Mukerjee ((1999a); Chapters 2, 6 & 7). Many of these results relate to situations where all factorial effects involving the same number of factors are considered equally important and, as such, the underlying model involves the general mean and all factorial effects involving up to a specified number of factors. In practice however, all factorial effects involving the same number of factors may not always be equally important and often, an experimenter is interested in estimating the general mean, all main effects and only a specified set of two-factor interactions, all other interactions being assumed negligible. The issue of estimability and optimality in situations of this kind has been addressed by Hedayat in the context of two-level factorials and, by Dey and Mukerjee (1999b) for arbitrary factorials including the asymmetric ones. Using finite projective geometry, Dey and Suen (2002) recently obtained several families of optimal plans under the stated model for symmetric factorials of the type m n , where m is a prime or a prime power. Continuing with this line of research, in this paper we obtain optimal fractional factorial plans for asymmetric (mixed level) factorials under a model that includes the mean, all main effects and a specified set of two-factor interactions. All other interactions are assumed to be negligible. Throughout, the optimality criterion considered is the universal optimality of Kiefer (1975); see also Sinha and Mukerjee (1982). In Section 2, concepts and results from a finite projective geometry are used to obtain optimal plans for asymmetric factorials, where the levels of the factors as also the …
منابع مشابه
Orthogonal Arrays
(say) is called symmetric, otherwise, the array is said to be asymmetric. Several methods of construction of symmetric as well as asymmetric OAs are available in the literature. Some important methods will be discussed here. One of the principal applications of the OAs is in the selection of level combinations for fractional factorial experiments. An OA of strength t is equivalent to an orthogo...
متن کاملInter-effect Orthogonality and Optimality in Hierarchical Models
The study of optimal fractional factorial plans has received considerable attention over the last two decades. The universal optimality of plans based on orthogonal arrays was shown by Cheng (1980) and Mukerjee (1982). Various extensions of this result have also been reported in the literature; see Mukerjee (1995) for a brief review. Most of these results relate to situations where all factoria...
متن کاملFractional Factorial Split-Plot Designs with Minimum Aberration and Maximum Estimation Capacity
Considering general prime or prime powered factorials, we give a nite projective geometric formulation for regular fractional factorial split-plot designs. This provides a uniied framework for such designs and facilitates their systematic study under the criteria of minimum aberration and minimum secondary aberration; the latter criterion is introduced to achieve ner discrimination. We investig...
متن کاملQuasi-orthogonal Arrays and Optimal Fractional Factorial Plans
Generalizing orthogonal arrays, a new class of arrays called quasiorthogonal arrays, are introduced and it is shown that fractional factorial plans represented by these arrays are universally optimal under a wide class of models. Some methods of construction of quasi-orthogonal arrays are also described.
متن کاملDesign and Analysis Approach for Factorials with Observable Noise Factors in Robust Design
Recent literature has proposed integrating observed noise variables into a factorial design structure as an alternative robust design strategy. Because noise factors are often difficult to control or randomize, observing their values may be useful. This research compares existing alternatives involving unreplicated fractional factorial designs and observable noise factors.
متن کامل