pith. sign in

arxiv: 2607.02027 · v1 · pith:FL76L7WJnew · submitted 2026-07-02 · 📊 stat.ME

Grouped Orthogonal Arrays from Orthogonal Arrays and Difference Schemes

Meixin Liu , Chunyan Wang , Guangzhou Chen , C. Devon Lin This is my paper

Pith reviewed 2026-07-03 07:52 UTC · model grok-4.3

classification 📊 stat.ME
keywords grouped orthogonal arraysorthogonal arraysdifference schemesexperimental designcomputer experimentsphysical experiments
0
0 comments X

The pith

Constructions combining orthogonal arrays and difference schemes produce grouped orthogonal arrays with significantly more groups and larger group sizes.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops several constructions for grouped orthogonal arrays using orthogonal arrays and difference schemes. These designs address needs in computer experiments with grouped inputs and physical experiments where cross-group interactions are negligible. By providing new arrays with larger numbers of groups and group sizes, the constructions expand the available options for efficient experimental designs. Sympathetic readers would care because these arrays allow for more flexible and larger-scale experiments without requiring full interaction modeling between groups.

Core claim

The authors show that grouped orthogonal arrays can be constructed from existing orthogonal arrays and difference schemes in ways that preserve the necessary orthogonality properties, resulting in a large collection of new designs with significantly larger numbers of groups and group sizes than previously available.

What carries the argument

Constructions that combine orthogonal arrays with difference schemes while preserving orthogonality for grouped factors.

If this is right

  • Designs can now accommodate more groups of factors in experiments.
  • Group sizes can be increased while maintaining the grouped orthogonal array properties.
  • Catalogs of such arrays for practical use in computer and physical experiments are expanded.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • These new constructions could be applied to other experimental settings involving grouped variables.
  • Connections to broader combinatorial constructions might yield even more designs.
  • Testing these arrays in actual computer experiments could reveal practical benefits in run efficiency.

Load-bearing premise

Suitable orthogonal arrays and difference schemes exist that can be combined while preserving the required orthogonality properties for the grouped arrays.

What would settle it

Constructing a specific grouped orthogonal array using the proposed method and verifying it does not satisfy the orthogonality conditions, or identifying parameters where no valid input arrays exist.

read the original abstract

Grouped orthogonal arrays were introduced to address experimental design problems arising in computer experiments with grouped inputs, as well as in physical experiments where interactions between factors from different groups are assumed to be negligible. Motivated by the growing need for flexible and efficient designs under such settings, this article develops several constructions to expand the existing catalogs of grouped orthogonal arrays. The proposed constructions provide a large collection of new grouped orthogonal arrays with significantly larger numbers of groups and group sizes.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript develops several constructions for grouped orthogonal arrays (GOAs) by combining orthogonal arrays and difference schemes. These constructions are claimed to yield a large collection of new GOAs with significantly larger numbers of groups and group sizes compared to existing ones, addressing needs in computer experiments with grouped inputs and physical experiments with negligible inter-group interactions.

Significance. If the constructions are valid and the claimed improvements in group numbers and sizes hold without hidden parameter restrictions, the work would substantially expand the catalog of usable GOAs. This is significant for experimental design applications where grouped factors appear and inter-group interactions can be ignored.

major comments (2)
  1. [Main construction theorem] The central construction (likely in the main theorem section following the abstract) asserts that suitable orthogonal arrays and difference schemes can always be combined while preserving the required orthogonality for the resulting GOA. However, the weakest assumption in the reader's note identifies this preservation step; the manuscript must provide an explicit proof or counterexample check that the strength parameters (e.g., strength 2 or 3) are maintained after the combination, as this is load-bearing for the claim of 'significantly larger' arrays.
  2. [Abstract and comparison tables] The abstract and any comparison tables claim 'significantly larger numbers of groups and group sizes,' but without explicit parameter tables or existence conditions (e.g., for given run size N, number of groups g, and group sizes), it is unclear whether the new arrays improve on known bounds or merely restate existing difference-scheme products. A concrete example with numerical parameters before and after construction is needed to substantiate the improvement.
minor comments (2)
  1. [Introduction] Notation for the grouped factors and the resulting array strength should be defined more explicitly in the introduction to avoid ambiguity with standard OA notation.
  2. Add a small worked example (e.g., a 12-run GOA) immediately after the first construction to illustrate the combination step.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on our constructions of grouped orthogonal arrays. We address the major comments point by point below.

read point-by-point responses

  1. Referee: [Main construction theorem] The central construction (likely in the main theorem section following the abstract) asserts that suitable orthogonal arrays and difference schemes can always be combined while preserving the required orthogonality for the resulting GOA. However, the weakest assumption in the reader's note identifies this preservation step; the manuscript must provide an explicit proof or counterexample check that the strength parameters (e.g., strength 2 or 3) are maintained after the combination, as this is load-bearing for the claim of 'significantly larger' arrays.

    Authors: We agree that the explicit verification of strength preservation after combining the orthogonal array and difference scheme is central to the result. The proof in the main theorem relies on the defining orthogonality properties of the inputs to establish that the output GOA inherits the required strength; however, we will expand this argument with additional intermediate steps to make the preservation explicit and easier to follow. revision: yes

  2. Referee: [Abstract and comparison tables] The abstract and any comparison tables claim 'significantly larger numbers of groups and group sizes,' but without explicit parameter tables or existence conditions (e.g., for given run size N, number of groups g, and group sizes), it is unclear whether the new arrays improve on known bounds or merely restate existing difference-scheme products. A concrete example with numerical parameters before and after construction is needed to substantiate the improvement.

    Authors: We accept that concrete numerical comparisons are needed to demonstrate the claimed improvements. The revised manuscript will add a table listing specific parameter sets (N, g, group sizes) for both prior constructions and the new ones obtained from our methods, together with at least one fully worked numerical example showing the increase in the number of groups and group sizes. revision: yes

Circularity Check

0 steps flagged

No significant circularity; constructions are explicit and self-contained

full rationale

The paper presents combinatorial constructions that build grouped orthogonal arrays directly from standard external objects (orthogonal arrays and difference schemes). No equations or steps reduce a claimed result to a fitted parameter, self-definition, or load-bearing self-citation chain; the derivations consist of explicit mappings that preserve required orthogonality and are independently verifiable from the input designs. This matches the most common honest finding for construction papers in design theory.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only abstract reviewed; no explicit free parameters, axioms, or invented entities can be identified.

pith-pipeline@v0.9.1-grok · 5596 in / 859 out tokens · 22096 ms · 2026-07-03T07:52:59.702509+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

49 extracted references · 2 canonical work pages

  1. [1]

    Applied Stochastic Models in Business and Industry , volume=

    Data quality: What if deming were born today? , author=. Applied Stochastic Models in Business and Industry , volume=. 2025 , publisher=

    2025

  2. [2]

    Statistica Sinica , pages=

    Generalized resolution and minimum aberration criteria for Plackett-Burman and other nonregular factorial designs , author=. Statistica Sinica , pages=. 1999 , publisher=

    1999

  3. [3]

    Statistica Sinica , pages=

    Minimum moment aberration for nonregular designs and supersaturated designs , author=. Statistica Sinica , pages=. 2003 , publisher=

    2003

  4. [4]

    The Annals of Statistics , volume=

    Generalized minimum aberration for asymmetrical fractional factorial designs , author=. The Annals of Statistics , volume=. 2001 , publisher=

    2001

  5. [5]

    Journal of the American Statistical Association , volume=

    Fast approximation of Shapley values through fractional factorial designs , author=. Journal of the American Statistical Association , volume=. 2026 , publisher=

    2026

  6. [6]

    Journal of the American Statistical Association , pages=

    Bayesian optimization for branching and nested hyperparameters in deep learning , author=. Journal of the American Statistical Association , pages=. 2026 , publisher=

    2026

  7. [7]

    The Annals of Statistics , volume=

    Construction of mixed orthogonal arrays with high strength , author=. The Annals of Statistics , volume=. 2021 , publisher=

    2021

  8. [8]

    Quality and Reliability Engineering International , volume=

    Small mixed-level orthogonal designs , author=. Quality and Reliability Engineering International , volume=. 2025 , publisher=

    2025

  9. [9]

    Wiley Interdisciplinary Reviews: Computational Statistics , volume=

    Orthogonal arrays: A review , author=. Wiley Interdisciplinary Reviews: Computational Statistics , volume=. 2025 , publisher=

    2025

  10. [10]

    Statistica Sinica , pages=

    Bayesian optimal fractional factorials , author=. Statistica Sinica , pages=. 2001 , publisher=

    2001

  11. [11]

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences , volume=

    On the construction of variable strength orthogonal arrays , author=. IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences , volume=. 2023 , publisher=

    2023

  12. [12]

    Designs, Codes and Cryptography , volume=

    A construction for strength-3 covering arrays from linear feedback shift register sequences , author=. Designs, Codes and Cryptography , volume=. 2014 , publisher=

    2014

  13. [13]

    2020 , publisher=

    Surrogates: Gaussian Process Modeling, Design, and Optimization for the Applied Sciences , author=. 2020 , publisher=

    2020

  14. [14]

    Structural and Multidisciplinary Optimization , volume=

    A comprehensive review of digital twin—part 1: modeling and twinning enabling technologies , author=. Structural and Multidisciplinary Optimization , volume=. 2022 , publisher=

    2022

  15. [15]

    SIAM/ASA Journal on Uncertainty Quantification , volume=

    Grouped orthogonal arrays for computer experiments with grouped inputs , author=. SIAM/ASA Journal on Uncertainty Quantification , volume=. 2025 , publisher=

    2025

  16. [16]

    2011 , publisher=

    Experiments: Planning, Analysis, and Optimization (Second Edition) , author=. 2011 , publisher=

    2011

  17. [17]

    2025 , publisher=

    Theory of Nonregular Factorial Designs , author=. 2025 , publisher=

    2025

  18. [18]

    2025 , howpublished =

    Complete series of non-isomorphic orthogonal arrays , author=. 2025 , howpublished =

    2025

  19. [19]

    2025 , howpublished =

    A Library of Orthogonal Arrays , author =. 2025 , howpublished =

    2025

  20. [20]

    Statistica Sinica , number =

    Designs of variable resolution robust to non-negligible two-factor interactions , author =. Statistica Sinica , number =. 2016 , publisher=

    2016

  21. [21]

    and Pang, S

    Zhang, Q. and Pang, S. and Li, Y. , TITLE =. IEICE Trans. Foundamentals , FJOURNAL =. 2023 , PAGES =

    2023

  22. [22]

    Designs, Codes and Cryptography , VOLUME =

    Raaphorst, Sebastian and Moura, Lucia and Stevens, Brett , TITLE =. Designs, Codes and Cryptography , VOLUME =. 2014 , NUMBER =

    2014

  23. [23]

    Bose, R. C. and Bush, K. A. , title =. The Annals of Mathematical Statistics , number =. 1952 , doi =

    1952

  24. [24]

    Bush, K. A. , journal =. Orthogonal Arrays of Index Unity , urldate =

  25. [25]

    1999 , publisher=

    Orthogonal Arrays: Theory and Applications , author=. 1999 , publisher=

    1999

  26. [26]

    , journal =

    Fries, Arthur and Hunter, William G. , journal =. Minimum Aberration 2k-p Designs , urldate =

  27. [27]

    A catalogue of three-level regular fractional factorial designs , volume =

    Xu, Hongquan , year =. A catalogue of three-level regular fractional factorial designs , volume =. Metrika , doi =

  28. [28]

    Construction of some series of orthogonal arrays , author=. Sankhy. 1981 , publisher=

    1981

  29. [29]

    Journal of Statistical Planning and Inference , volume=

    On difference schemes and orthogonal arrays of strength t , author=. Journal of Statistical Planning and Inference , volume=. 1996 , publisher=

    1996

  30. [30]

    MINIMUM MOMENT ABERRATION FOR NONREGULAR DESIGNS AND SUPERSATURATED DESIGNS , urldate =

    Xu, Hongquan , journal =. MINIMUM MOMENT ABERRATION FOR NONREGULAR DESIGNS AND SUPERSATURATED DESIGNS , urldate =

  31. [31]

    Lin, C. D. , title = ". Biometrika , volume =

  32. [32]

    Devon , journal =

    Lekivetz, Ryan and Lin, C. Devon , journal =. DESIGNS OF VARIABLE RESOLUTION ROBUST TO NON-NEGLIGIBLE TWO-FACTOR INTERACTIONS , urldate =

  33. [33]

    Journal of the American Statistical Association , volume =

    Tang, Boxin , title =. Journal of the American Statistical Association , volume =. 1993 , publisher =. doi:10.1080/01621459.1993.10476423 , URL =

    work page doi:10.1080/01621459.1993.10476423 1993

  34. [34]

    ACM SIGACT News , volume=

    Combinatorial designs: constructions and analysis , author=. ACM SIGACT News , volume=. 2008 , publisher=

    2008

  35. [35]

    He, Y. Z. and Lin, C. D. and Sun, F. , title =. The Annals of Statistics , number =

  36. [36]

    A construction for strength-3 covering arrays from linear feedback shift register sequences , volume =

    Raaphorst, Sebastian and Moura, Lucia and Stevens, Brett , year =. A construction for strength-3 covering arrays from linear feedback shift register sequences , volume =

  37. [37]

    Metrika , volume=

    A catalogue of three-level regular fractional factorial designs , author=. Metrika , volume=. 2005 , publisher=

    2005

  38. [38]

    Canadian Journal of Mathematics , author=

    Partitioning Projective Geometries into Caps , volume=. Canadian Journal of Mathematics , author=. 1985 , pages=. doi:10.4153/CJM-1985-063-1 , number=

    work page doi:10.4153/cjm-1985-063-1 1985

  39. [39]

    The Annals of Mathematical Statistics , volume=

    On the problem of construction of orthogonal arrays , author=. The Annals of Mathematical Statistics , volume=. 1954 , publisher=

    1954

  40. [40]

    McKay, M. D. and Beckman, R. J. and Conover, W. J. , journal =. 1979 , title =

    1979

  41. [41]

    1996 , title =

    Box, George and Tyssedal, John , journal =. 1996 , title =

    1996

  42. [42]

    Choudhury, Piyasi , year =

  43. [43]

    Xu, Hongquan and Phoa, Frederick K. H. and Wong, Weng Kee , year =

  44. [44]

    2013 , title =

    Kharaghani, Hadi and Tayfeh-Rezaie, Behruz , journal =. 2013 , title =

    2013

  45. [45]

    2023 , title =

    Shi, Chenlu and Tang, Boxin , journal =. 2023 , title =

    2023

  46. [46]

    and He, Y

    Chen, G. and He, Y. and Lin, C. D. and Sun, F. , journal =. 2024 , volume =

    2024

  47. [47]

    , journal =

    Kuhfeld, Warren F. , journal =. 2010 , title =

    2010

  48. [48]

    Eendebak, P. T. and Vazquez, A. R. , journal =. 2019 , doi =

    2019

  49. [49]

    2009 , title =

    Xu, Hongquan , journal =. 2009 , title =

    2009