A construction method for orthogonal latin hypercube designs with prime power levels

Fang Pang, Min Qian Liu, Dennis K.J. Lin

Research output: Contribution to journalArticle

39 Citations (Scopus)

Abstract

Latin hypercube design (LHD) is popularly used in designing computer experiments. This paper explores how to construct LHDs with pd (d = 2c) runs and up to (pd - l)/(p - 1) factors in which all main effects are orthogonal. This is accomplished by rotating groups of factors in a pd-run regular saturated factorial design. These rotated factorial designs are easy to construct and preserve many attractive properties of standard factorial designs. The proposed method covers the one by Steinberg and Lin (2006) as a special case and is able to generate more orthogonal LHDs with attractive properties. Theoretical properties as well as the construction algorithm are discussed, with an example for illustration.

Original languageEnglish (US)
Pages (from-to)1721-1728
Number of pages8
JournalStatistica Sinica
Volume19
Issue number4
StatePublished - Oct 2009

Fingerprint

Latin Hypercube Design
Orthogonal Design
Factorial Design
Saturated Design
Computer Experiments
Main Effect
Rotating
Cover
Factorial design
Factors

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{717c718811824bb68205f33dda86a72a,
title = "A construction method for orthogonal latin hypercube designs with prime power levels",
abstract = "Latin hypercube design (LHD) is popularly used in designing computer experiments. This paper explores how to construct LHDs with pd (d = 2c) runs and up to (pd - l)/(p - 1) factors in which all main effects are orthogonal. This is accomplished by rotating groups of factors in a pd-run regular saturated factorial design. These rotated factorial designs are easy to construct and preserve many attractive properties of standard factorial designs. The proposed method covers the one by Steinberg and Lin (2006) as a special case and is able to generate more orthogonal LHDs with attractive properties. Theoretical properties as well as the construction algorithm are discussed, with an example for illustration.",
author = "Fang Pang and Liu, {Min Qian} and Lin, {Dennis K.J.}",
year = "2009",
month = "10",
language = "English (US)",
volume = "19",
pages = "1721--1728",
journal = "Statistica Sinica",
issn = "1017-0405",
publisher = "Institute of Statistical Science",
number = "4",

}

A construction method for orthogonal latin hypercube designs with prime power levels. / Pang, Fang; Liu, Min Qian; Lin, Dennis K.J.

In: Statistica Sinica, Vol. 19, No. 4, 10.2009, p. 1721-1728.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A construction method for orthogonal latin hypercube designs with prime power levels

AU - Pang, Fang

AU - Liu, Min Qian

AU - Lin, Dennis K.J.

PY - 2009/10

Y1 - 2009/10

N2 - Latin hypercube design (LHD) is popularly used in designing computer experiments. This paper explores how to construct LHDs with pd (d = 2c) runs and up to (pd - l)/(p - 1) factors in which all main effects are orthogonal. This is accomplished by rotating groups of factors in a pd-run regular saturated factorial design. These rotated factorial designs are easy to construct and preserve many attractive properties of standard factorial designs. The proposed method covers the one by Steinberg and Lin (2006) as a special case and is able to generate more orthogonal LHDs with attractive properties. Theoretical properties as well as the construction algorithm are discussed, with an example for illustration.

AB - Latin hypercube design (LHD) is popularly used in designing computer experiments. This paper explores how to construct LHDs with pd (d = 2c) runs and up to (pd - l)/(p - 1) factors in which all main effects are orthogonal. This is accomplished by rotating groups of factors in a pd-run regular saturated factorial design. These rotated factorial designs are easy to construct and preserve many attractive properties of standard factorial designs. The proposed method covers the one by Steinberg and Lin (2006) as a special case and is able to generate more orthogonal LHDs with attractive properties. Theoretical properties as well as the construction algorithm are discussed, with an example for illustration.

UR - http://www.scopus.com/inward/record.url?scp=71249122802&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=71249122802&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:71249122802

VL - 19

SP - 1721

EP - 1728

JO - Statistica Sinica

JF - Statistica Sinica

SN - 1017-0405

IS - 4

ER -