Optimal Blocked and Split-Plot Designs Ensuring Precise Pure-Error Estimation of the Variance Components

Abstract Textbooks on response surface methodology generally stress the importance of lack-of-fit tests and estimation of pure error. For lack-of-fit tests to be possible and other inference to be unbiased, experiments should allow for pure-error estimation. Therefore, they should involve replicated treatments. While most textbooks focus on lack-of-fit testing in the context of completely randomized designs, many response surface experiments are not completely randomized and require block or split-plot structures. The analysis of data from blocked or split-plot experiments is generally based on a mixed regression model with two variance components instead of one. In this article, we present a novel approach to designing blocked and split-plot experiments which ensures that the two variance components can be efficiently estimated from pure error and guarantees a precise estimation of the response surface model. Our novel approach involves a new Bayesian compound D-optimal design criterion which pays attention to both the variance components and the fixed treatment effects. One part of the compound criterion (the part concerned with the treatment effects) is based on the response surface model of interest, while the other part (which is concerned with pure-error estimates of the variance components) is based on the full treatment model. We demonstrate that our new criterion yields split-plot designs that outperform existing designs from the literature both in terms of the precision of the pure-error estimates and the precision of the estimates of the factor effects.

[1]  Connie M. Borror,et al.  Modifying a central composite design to model the process mean and variance when there are hard‐to‐change factors , 2006 .

[2]  R. K. Meyer,et al.  The Coordinate-Exchange Algorithm for Constructing Exact Optimal Experimental Designs , 1995 .

[3]  Anthony C. Atkinson,et al.  Optimum Experimental Designs, with SAS , 2007 .

[4]  Kalliopi Mylona,et al.  Optimal Design of Blocked and Split-Plot Experiments for Fixed Effects and Variance Component Estimation , 2014, Technometrics.

[5]  Douglas C. Montgomery,et al.  Response Surface Designs within a Split-Plot Structure , 2005 .

[6]  $$\mathcal{D}$$-optimal Minimum Support Mixture Designs in Blocks , 2006 .

[7]  Raymond H. Myers,et al.  Response Surface Methods for Bi-Randomization Structures , 1996 .

[8]  C. Anderson‐Cook.  Response Surfaces, Mixtures, and Ridge Analyses , 2008 .

[9]  Kalliopi Mylona,et al.  Optimum design of experiments for statistical inference : discussion , 2012 .

[10]  G. Geoffrey Vining,et al.  Classes of Split‐plot Response Surface Designs for Equivalent Estimation , 2006, Qual. Reliab. Eng. Int..

[11]  Peter Goos,et al.  Staggered-Level Designs for Experiments With More Than One Hard-to-Change Factor , 2012, Technometrics.

[12]  Steven G. Gilmour,et al.  Some practical advice on polynomial regression analysis from blocked response surface designs , 2000 .

[13]  Henry W. Altland.  Experiments: Planning, Analysis, and Parameter Design Optimization , 2001, Technometrics.

[14]  P. Goos,et al.  Quadrature Methods for Bayesian Optimal Design of Experiments With Nonnormal Prior Distributions , 2018 .

[15]  A. Donev,et al.  D-optimal minimum support mixture designs in blocks Goos , 2017 .

[16]  Peter Goos,et al.  Practical Inference from Industrial Split-Plot Designs , 2006 .

[17]  Peter Goos,et al.  Comparing different sampling schemes for approximating the integrals involved in the efficient design of stated choice experiments , 2010 .

[18]  Steven G. Gilmour,et al.  Split-Plot and Multi-Stratum Designs for Statistical Inference , 2017, Technometrics.

[19]  Peter Goos,et al.  A candidate-set-free algorithm for generating D-optimal split-plot designs , 2007, Journal of the Royal Statistical Society. Series C, Applied statistics.

[20]  Steven G. Gilmour,et al.  Optimum design of experiments for statistical inference , 2012 .

[21]  T. Robinson,et al.  DP-Optimality in Terms of Multiple Criteria and Its Application to the Split-Plot Design , 2017 .

[22]  W. Näther.  Optimum experimental designs , 1994 .

[23]  S. Gilmour,et al.  Testing for Lack of Fit in Blocked, Split-Plot, and Other Multi-Stratum Designs , 2017 .

[24]  K. Chaloner,et al.  Optimal Bayesian design applied to logistic regression experiments , 1989 .

[25]  M. Bliemer,et al.  Approximation of bayesian efficiency in experimental choice designs , 2008 .

[26]  Optimal Designs for Nonlinear Models with Random Block Effects , 2018 .

[27]  Peter Goos,et al.  Tailor-Made Split-Plot Designs for Mixture and Process Variables , 2007 .

[28]  Peter Goos,et al.  Optimal versus orthogonal and equivalent‐estimation design of blocked and split‐plot experiments , 2006 .

[29]  G. Geoffrey Vining,et al.  Split-Plot Designs and Estimation Methods for Mixture Experiments With Process Variables , 2002, Technometrics.