Optimal mixed-level supersaturated design

Kai Tai Fang; K. J. Lin Dennis; Min Qian Liu

doi:10.1007/s001840300266

Optimal mixed-level supersaturated design

Kai Tai Fang, K. J. Lin Dennis^*, Min Qian Liu

^*Corresponding author for this work

Research output: Contribution to journal › Journal article › peer-review

112 Citations (Scopus)

12 Downloads (Pure)

Abstract

A supersaturated design is essentially a fractional factorial in which the number of potential effects is greater than the number of runs. In this paper, E(f_NOD) criterion is employed for comparing supersaturated designs from the viewpoint of orthogonality and uniformity, and a lower bound of E(f_NOD) which can serve as a benchmark of design optimality is obtained. It is shown that the existing E(s²) and ave x² criteria (for two- and three-level supersaturated designs respectively) are in fact special cases of this criterion. Furthermore, a construction method for mixed-level supersaturated designs is proposed and some properties of the resulting designs are investigated.

Original language	English
Pages (from-to)	279-291
Number of pages	13
Journal	Metrika
Volume	58
Issue number	3
DOIs	https://doi.org/10.1007/s001840300266
Publication status	Published - Dec 2003
Externally published	Yes

Scopus Subject Areas

Statistics and Probability
Statistics, Probability and Uncertainty

User-Defined Keywords

Discrepancy
Hamming distance
Orthogonal array
Supersatu-rated design
U-type design
Uniformity

Access to Document

10.1007/s001840300266

Full TextAccepted author manuscript, 9.4 MB

Cite this

@article{87fd7250e1fc4badb2c84fd6745ab1d5,

title = "Optimal mixed-level supersaturated design",

abstract = "A supersaturated design is essentially a fractional factorial in which the number of potential effects is greater than the number of runs. In this paper, E(fNOD) criterion is employed for comparing supersaturated designs from the viewpoint of orthogonality and uniformity, and a lower bound of E(fNOD) which can serve as a benchmark of design optimality is obtained. It is shown that the existing E(s2) and ave x2 criteria (for two- and three-level supersaturated designs respectively) are in fact special cases of this criterion. Furthermore, a construction method for mixed-level supersaturated designs is proposed and some properties of the resulting designs are investigated.",

keywords = "Discrepancy, Hamming distance, Orthogonal array, Supersatu-rated design, U-type design, Uniformity",

author = "Fang, {Kai Tai} and {Lin Dennis}, {K. J.} and Liu, {Min Qian}",

year = "2003",

month = dec,

doi = "10.1007/s001840300266",

language = "English",

volume = "58",

pages = "279--291",

journal = "Metrika",

issn = "0026-1335",

publisher = "Springer Verlag",

number = "3",

}

TY - JOUR

T1 - Optimal mixed-level supersaturated design

AU - Fang, Kai Tai

AU - Lin Dennis, K. J.

AU - Liu, Min Qian

PY - 2003/12

Y1 - 2003/12

N2 - A supersaturated design is essentially a fractional factorial in which the number of potential effects is greater than the number of runs. In this paper, E(fNOD) criterion is employed for comparing supersaturated designs from the viewpoint of orthogonality and uniformity, and a lower bound of E(fNOD) which can serve as a benchmark of design optimality is obtained. It is shown that the existing E(s2) and ave x2 criteria (for two- and three-level supersaturated designs respectively) are in fact special cases of this criterion. Furthermore, a construction method for mixed-level supersaturated designs is proposed and some properties of the resulting designs are investigated.

AB - A supersaturated design is essentially a fractional factorial in which the number of potential effects is greater than the number of runs. In this paper, E(fNOD) criterion is employed for comparing supersaturated designs from the viewpoint of orthogonality and uniformity, and a lower bound of E(fNOD) which can serve as a benchmark of design optimality is obtained. It is shown that the existing E(s2) and ave x2 criteria (for two- and three-level supersaturated designs respectively) are in fact special cases of this criterion. Furthermore, a construction method for mixed-level supersaturated designs is proposed and some properties of the resulting designs are investigated.

KW - Discrepancy

KW - Hamming distance

KW - Orthogonal array

KW - Supersatu-rated design

KW - U-type design

KW - Uniformity

UR - http://www.scopus.com/inward/record.url?scp=1342338205&partnerID=8YFLogxK

U2 - 10.1007/s001840300266

DO - 10.1007/s001840300266

M3 - Journal article

AN - SCOPUS:1342338205

SN - 0026-1335

VL - 58

SP - 279

EP - 291

JO - Metrika

JF - Metrika

IS - 3

ER -

Optimal mixed-level supersaturated design

Abstract

Scopus Subject Areas

User-Defined Keywords

Access to Document

Other files and links

Fingerprint

Cite this