Designing Optimal Large Four-Level Experiments: A New Technique Without Recourse to Optimization Softwares

Abstract

Experimental design is an effective statistical tool that is extensively applied in modern industry, engineering, and science. It is proved that experimental design is a powerful and efficient means to screen the relationships between input factors and their responses, and to distinguish significant and unimportant factor effects. In many practical situations, experimenters are faced with large experiments having four-level factors. Even though there are several techniques provided to design such experiments, the challenge faced by the experimenters is still daunting. The practice has demonstrated that the existing techniques are highly time-consuming optimization procedures, satisfactory outcomes are not guaranteed, and non-mathematicians face a significant challenge in dealing with them. A new technique that can overcome these defects of the existing techniques is presented in this paper. The results demonstrated that the proposed technique outperformed the current techniques in terms of construction simplicity, computational efficiency and achieving satisfactory results capability. For non-mathematician experimenters, the new technique is much easier and simpler than the current techniques, as it allows them to design optimal large experiments without the recourse to optimization softwares. The optimality is discussed from four basic perspectives: maximizing the dissimilarity among experimental runs, maximizing the number of independent factors, minimizing the confounding among factors, and filling the experimental domain uniformly with as few gaps as possible.

Keywords

Multiple quadrupling technique / TA algorithm / Augmented design technique / Level permutation technique / Confounding / Hamming distance / Space-filling

References

[1] Ankenman BE. Design of experiments with two- and four-level factors. J. Quality Technol.. 1999, 31 4 363-375

[2] Chen H, Cheng CS. Doubling and projection: a method of constructing two-level designs of resolution IV. Ann. Stat.. 2006, 34 546-558

[3] Cheng CS, Steinberg DM, Sun DX. Minimum aberration and model robustness for two-level fractional factorial designs. J. R. Stat. Soc. Ser. B. 1999, 61 85-93

[4] Cheng CS, Deng LY, Tang B. Generalized minimum aberration and design efficiency for nonregular fractional factorial designs. Stat. Sin.. 2002, 12 991-1000

[5] Clark JB, Dean AM. Equivalence of fractional factorial designs. Stat. Sin.. 2001, 11 537-547

[6] Davies OL, Hay WA. The construction and uses of fractional factorial designs in industrial research. Biometrics. 1950, 6 3 233-249

[7] Dey A, Mukerjee R. Fractional Factorial Plans. 1999 New York: Wiley

[8] Edmondson RN. Agricultural response surface experiments in view of four-level factorial designs. Biometrics. 1991, 47 4 1435-1448

[9] Elsawah AM. An appealing technique for designing optimal large experiments with three-level factors. J. Comput. Appl. Math.. 2021, 384 113164

[10] Elsawah AM. Building some bridges among various experimental designs. J. Korean Stat. Soc.. 2020, 49 55-81

[11] Elsawah AM. Constructing optimal router bit life sequential experimental designs: new results with a case study. Commun. Stat. Simul. Comput.. 2019, 48 3 723-752

[12] Elsawah AM. Choice of optimal second stage designs in two-stage experiments. Comput. Stat.. 2018, 33 2 933-965

[13] Elsawah AM. A powerful and efficient algorithm for breaking the links between aliased effects in asymmetric designs. Aust. N. Z. J. Stat.. 2017, 59 1 17-41

[14] Elsawah AM. Constructing optimal asymmetric combined designs via Lee discrepancy. Stat. Prob. Lett.. 2016, 118 24-31

[15] Elsawah AM, Fang KT. A catalog of optimal foldover plans for constructing U-uniform minimum aberration four-level combined designs. J. Appl. Stat.. 2019, 46 7 1288-1322

[16] Elsawah AM, Fang KT. New results on quaternary codes and their Gray map images for constructing uniform designs. Metrika. 2018, 81 3 307-336

[17] Elsawah AM, Fang KT, Ke X. New recommended designs for screening either qualitative or quantitative factors. Stat. Pap.. 2019

[18] Elsawah AM, Tang Y, Fang KT. Constructing optimal projection designs. Statistics. 2019, 53 6 1357-1385

[19] Fang KT. The uniform designs: application of number-theoretic methods in experimental design. Acta Math. Appl. Sin.. 1980, 3 363-372

[20] Fang KT, Ke X, Elsawah AM. Construction of uniform designs via an adjusted threshold accepting algorithm. J. Complex.. 2017, 43 28-37

[21] Fang KT, Lin DKJ, Winker P, Zhang Y. Uniform design: theory and application. Technometrics. 2000, 42 237-248

[22] Fang KT, Lu X, Winker P. Lower bounds for centered and wrap-around $L_2$-discrepancies and construction of uniform designs by threshold accepting. J. Complex.. 2003, 19 692-711

[23] Fang KT, Ma CX, Mukerjee R. Fang KT, Hickernell FJ, Niederreiter H. Uniformity in fractional factorials. Monte Carlo and Quasi-Monte Carlo Methods in Scientific Computing. 2002 Berlin: Springer

[24] Fang KT, Hickernell FJ. The uniform design and its applications. Bull. Inst. Int. Stat.. 1995, 1 333-349

[25] Fries A, Hunter WG. Minimum Aberration $2^{k-p}$ Designs. Technometrics. 1980, 22 601-608

[26] Hedayat AS, Sloane NJ, Stufken J. Orthogonal Arrays: Theory and Application. 1999 Berlin: Springer

[27] Hickernell FJ. A generalized discrepancy and quadrature error bound. Math. Comput.. 1998, 67 299-322

[28] Hickernell, F.J.: Lattice rules: how well do they measure up? In: Hellekalek, P., Larcher, G. (Eds.), Random and Quasi-Random Point Sets. In: Lecture Notes in Statistics, vol. 138. Springer, New York, pp. 109–166 (1998b)

[29] Kirkpatrick S, Gelatt CD, Vecchi MP. Optimization by simulated annealing. Science. 1983, 220 4598 671-680

[30] Koehl V, Parizet E. Influence of structural variability upon sound perception: usefulness of fractional factorial designs. Appl. Acoust.. 2006, 67 3 249-270

[31] Lan WG, Wong MK, Chen N, Sin YM. Four-level orthogonal array design as a chemometric approach to the optimization of polarographic reaction system for phosphorus determination. Talanta. 1994, 41 11 1917-1927

[32] Li W, Lin DKJ, Ye KQ. Optimal foldover plans for non-regular orthogonal designs. Technometrics. 2003, 45 347-351

[33] Li H, Qin H. Quadrupling: construction of uniform designs with large run sizes. Metrika. 2020, 83 527-544

[34] Liu MQ, Fang KT, Hickernell FJ. Connection among different criteria for asymmetrical fractional factorial designs. Stat. Sin.. 2006, 16 1285-297

[35] Lu, X., Fang, K.T., Xu, Q.F., Yin, J.X.: Balance pattern and BP-optimal factorial designs. Technical Report-324, Hong Kong Baptist University (2002)

[36] Ma CX, Fang KT. A note on generalized aberration in factorial designs. Metrika. 2001, 53 85-93

[37] Phadke MS. Design optimization case studies. AT & T Techn. J.. 1986, 65 51-68

[38] Phoa FKH. A code arithmetic approach for quaternary code designs and its application to $(1/64)\text{th}$-fractions. Ann. Stat.. 2012, 40 6 3161-3175

[39] Phoa FKH, Xu H. Quarter-fraction factorial designs constructed via quaternary codes. Ann. Stat.. 2009, 37 2561-2581

[40] Phoa FKH, Xu H, Wong WK. The use of nonregular fractional factorial designs in combination toxicity studies. Food Chem. Toxicol.. 2009, 47 9 2183-2188

[41] Rytz, et al.: Using fractional factorial designs with mixture constraints to improve nutritional value and sensory properties of processed food. Food Quality Pref. 58, 71–75 (2017)

[42] Tang B, Deng LY. Minimum $G_2$-aberration for non-regular fractional factorial designs. Ann. Stat.. 1999, 27 1914-1926

[43] Tang Y, Xu H. An effective construction method for multi-level uniform designs. J. Stat. Plan Inference. 2013, 143 1583-1589

[44] Wang Y, Fang KT. A not on uniform distribution and experimental design. Chin. Sci. Bull.. 1981, 26 485-489

[45] Wang B, Robert GM, John FB. A note on the selection of optimal foldover plans for 16- and 32-run fractional factorial designs. J. Stat. Plan. Inference. 2010, 140 1497-1500

[46] Wakjira TG, Nehdi ML, Ebead U. Fractional factorial design model for seismic performance of RC bridge piers retrofitted with steel-reinforced polymer composites. Eng. Struct.. 2020, 221 15 111100

[47] Winker P, Fang KT. Niederreiter H, Hellekalek P, Larcher G, Zinterhof P. Optimal U-type designs. Monte Carlo and Quasi-Monte Carlo Methods. 1997 New York: Springer. 436-488

[48] Xu H. Minimum moment aberration for nonregular designs and supersaturated designs. Stat. Sin.. 2003, 13 691-708

[49] Xu H, Wu CFJ. Generalized minimum aberration for asymmetrical fractional factorial designs. Ann. Stat.. 2001, 29 549-560

[50] Xu G, Zhang J, Tang Y. Level permutation method for constructing uniform designs under the wrap-around $L_2$-discrepancy. J. Complex.. 2014, 30 46-53

[51] Yang F, Zhou Y-D, Zhang X-R. Augmented uniform designs. J. Stat. Plan. Inference. 2017, 182 64-737

[52] Zhang A, Fang KT, Li R, Sudjianto A. Majorization framework for balanced lattice designs. Ann. Stat.. 2005, 33 6 2837-2853

[53] Zhou YD, Xu H. Space-filling fractional factorial designs. J. Am. Stat. Assoc.. 2014, 109 507 1134-1144

[54] Zhou YD, Ning JH, Song XB. Lee discrepancy and its applications in experimental designs. Stat. Prob. Lett.. 2008, 78 1933-1942

Funding

UIC(R201810)

UIC(R202010)