Abstract

Designs for computer experiments are an important class of experimen

tal designs that offer significant cost and time savings over physical experiments.

In this paper, we introduce the concept of sliced orthogonal designs for computer

experiments, a generalisation of sliced Latin hypercube designs.

We propose

methods for constructing these sliced orthogonal designs, which form a special

class suitable for both first-order and second-order models, with each slice constituting an orthogonal sub-design. The construction methods leverage known

sequences with zero autocorrelation function, such as T-sequences and Golay sequences, as well as disjoint amicable sequences. This approach introduces, for the

first time, infinite families of such designs. The generated designs are evaluated

using various criteria from the literature, and the results are presented in tables

for practitioners’ reference.

Information

Preprint No.SS-2024-0379
Manuscript IDSS-2024-0379
Complete AuthorsOmar A. Alhelali, S.D. Georgiou, S. Stylianou
Corresponding AuthorsS.D. Georgiou
Emailsstelios.georgiou@rmit.edu.au

References

  1. Alhelali, O. A., S. D. Georgiou, C. Koukouvinos, and S. Stylianou (2025). Orthogonal designs for computer experiments constructed from sequences with zero autocorrelation. Applied Numerical Mathematics 208(Part A), 22–31. Published online 2023; in print 2025.
  2. Bingham, D., R. R. Sitter, and B. Tang (2009). Orthogonal and nearly orthogonal designs for computer experiments. Biometrika 96(1), 51–65.
  3. Cao, R.-Y. and M.-Q. Liu (2015). Construction of second-order orthogonal sliced latin hypercube designs. Journal of Complexity 31(5), 762–772.
  4. Chen, P.-H. A., T. J. Santner, and A. M. Dean (2018). Sequential pareto minimization of physical systems using calibrated computer simulators. Statistica Sinica 28(2), 671–692.
  5. Cohen, G., D. Rubie, J. Seberry, C. Koukouvinos, S. Kounias, and M. Yamada (1989). A survey of base sequences, disjoint complementary sequences and od(4t; t, t, t, t). J. Comb. Math. Comb. Comput. 5, 69–104.
  6. Colbourn, C. J. and J. H. Dinitz (2006). Handbook of Combinatorial Designs, Second Edition (Discrete Mathematics and Its Applications). Chapman & Hall/CRC.
  7. Cooper, J. and J. S. Wallis (1972). A construction for hadamard arrays. Bulletin of the Australian Mathematical Society 7(2), 269–278.
  8. Djokovic, D. Z. (2010a). A new yang number and consequences. Designs, Codes and Cryptography 54, 201–204.
  9. Djokovic, D. Z. (2010b). Small orders of hadamard matrices and base sequences. arXiv preprint arXiv:1008.2043.
  10. Fang, K.-T., R. Li, and A. Sudjianto (2005). Design and modeling for computer experiments. CRC press.
  11. Fang, K.-T., D. K. Lin, and H. Qin (2003). A note on optimal foldover design. Statistics & probability letters 62(3), 245–250.
  12. Georgiou, S., C. Koukouvinos, and J. Seberry (2002). Short amicable sets. International Journal of Applied Mathematics 9, 161–187.
  13. Georgiou, S. D. (2011). Orthogonal designs for computer experiments. Journal of Statistical Planning and Inference 141(4), 1519–1525.
  14. Georgiou, S. D. and I. Efthimiou (2014). Some classes of orthogonal latin hypercube designs. Statistica Sinica 24(1), 101–120.
  15. Guo, B., X.-R. Li, M.-Q. Liu, and X. Yang (2023). Construction of orthogonal general sliced latin hypercube designs. Statistical Papers 64(3), 987–1014.
  16. He, Y. and B. Tang (2013). Strong orthogonal arrays and associated latin hypercubes for computer experiments. Biometrika 100(1), 254–260.
  17. Huang, H., J.-F. Yang, and M.-Q. Liu (2014). Construction of sliced (nearly) orthogonal latin hypercube designs. Journal of Complexity 30(3), 355–365.
  18. Huang, H., H. Yu, M.-Q. Liu, and D. Wu (2021). Construction of uniform designs and complexstructured uniform designs via partitionable t-designs. Statistica Sinica 31(4), 1689–1706.
  19. Joseph, V. R. and Y. Hung (2008). Orthogonal-maximin latin hypercube designs. Statistica Sinica 18(1), 171–186.
  20. Kharaghani, H. (2000). Arrays for orthogonal designs. Journal of Combinatorial Designs 8(3), 166–173.
  21. Kumar, A. A., B. N. Mandal, R. Parsad, S. Dash, and M. Kumar (2024). On construction of sliced orthogonal latin hypercube designs. Journal of Statistical Theory and Practice 18(4), 52.
  22. Li, W., M.-Q. Liu, and B. Tang (2021). A method of constructing maximin distance designs. Biometrika 108(4), 845–855.
  23. Lin, C. D. and B. Tang (2022). Latin hypercubes and space-filling designs. arXiv preprint arXiv:2203.06334.
  24. McKay, M. D., R. J. Beckman, and W. J. Conover (1979). A comparison of three methods for selecting values of input variables in the analysis of output from a computer code. Technometrics 42(1), 55–61.
  25. Morris, M. D. and T. J. Mitchell (1995). Exploratory designs for computational experiments. Journal of statistical planning and inference 43(3), 381–402.
  26. Qian, P. Z. (2012). Sliced latin hypercube designs. Journal of the American Statistical Association 107(497), 393–399.
  27. Santner, T. J., B. J. Williams, W. I. Notz, and B. J. Williams (2003). The design and analysis of computer experiments, Volume 1. Springer.
  28. Seberry, J. (2017, 01). Orthogonal Designs: Hadamard Matrices, Quadratic Forms and Algebras. Cham: Springer.
  29. Seberry, J. and M. Yamada (1992). Hadamard matrices, sequences and block designs. In J. H. Dinitz and D. R. Stinson (Eds.), Contemporary Design Theory – A Collection of Surveys, pp. 431–560. New York: John Wiley and Sons.
  30. Shi, C. and B. Tang (2020). Construction results for strong orthogonal arrays of strength three. Bernoulli 26(1), 418 – 431.
  31. Steinberg, D. M. and D. K. Lin (2006). A construction method for orthogonal latin hypercube designs. Biometrika 93(2), 279–288.
  32. Sun, F. and B. Tang (2017). A general rotation method for orthogonal latin hypercubes. Biometrika 104(2), 465–472.
  33. Tang, B. (1993). Orthogonal array-based latin hypercubes. Journal of the American statistical association 88(424), 1392–1397.
  34. Wang, C., J. Yang, and M.-Q. Liu (2022). Construction of strong group-orthogonal arrays. Statistica Sinica 32(3), 1–19.
  35. Wang, L., F. Sun, D. K. Lin, and M.-Q. Liu (2018). Construction of orthogonal symmetric latin hypercube designs. Statistica Sinica 28(3), 1503–1520.
  36. Wang, L., Q. Xiao, and H. Xu (2018). Optimal maximin L1-distance Latin hypercube designs based on good lattice point designs. The Annals of Statistics 46(6B), 3741 – 3766.
  37. Wang, X.-L., Y.-N. Zhao, J.-F. Yang, and M.-Q. Liu (2017). Construction of (nearly) orthogonal sliced latin hypercube designs. Statistics & Probability Letters 125, 174–180.
  38. Xiao, Q. and H. Xu (2018). Construction of maximin distance designs via level permutation and expansion. Statistica Sinica 28(3), 1395–1414.
  39. Yang, J., H. Chen, D. K. J. Lin, and M.-Q. Liu (2016). Construction of sliced maximinorthogonal latin hypercube designs. Statistica Sinica 26(2), 589–603.
  40. Yang, J.-F., C. D. Lin, P. Z. G. Qian, and D. K. J. Lin (2013). Construction of sliced orthogonal latin hypercube designs. Statistica Sinica 23(3), 1117–1130.
  41. Zhou, Y. and B. Tang (2019). Column-orthogonal strong orthogonal arrays of strength two plus and three minus. Biometrika 106(4), 997–1004. Omar A. Alhelali Mathematical Sciences, School of Science, RMIT University, Melbourne, VIC 3001, Australia.

Acknowledgments

The authors thank the editor, associate editor, and reviewers for their valuable comments and suggestions, which have helped improve the quality of

this article.