On the isomorphism of fractional factorial designs.

*(English)*Zbl 0979.62055Summary: Two fractional factorial designs are called isomorphic if one can be obtained from the other by relabeling the factors, reordering the runs, and switching the levels of factors. To identify the isomorphism of two \(s\)-factor \(n\)-run designs is known to be an NP hard problem, when \(n\) and \(s\) increase. There is no tractable algorithm for the identification of isomorphic designs.

We propose a new algorithm based on the centered \(L_2\)-discrepancy, a measure of uniformity, for detecting the isomorphism of fractional factorial designs. It is shown that the new algorithm is highly reliable and can significantly reduce the complexity of the computation. Theoretical justification for such an algorithm is also provided. The efficiency of the new algorithm is demonstrated by using several examples that have previously been discussed by many others.

We propose a new algorithm based on the centered \(L_2\)-discrepancy, a measure of uniformity, for detecting the isomorphism of fractional factorial designs. It is shown that the new algorithm is highly reliable and can significantly reduce the complexity of the computation. Theoretical justification for such an algorithm is also provided. The efficiency of the new algorithm is demonstrated by using several examples that have previously been discussed by many others.

PDF
BibTeX
XML
Cite

\textit{C.-X. Ma} et al., J. Complexity 17, No. 1, 86--97 (2001; Zbl 0979.62055)

Full Text:
DOI

**OpenURL**

##### References:

[1] | Bates, R.A.; Buck, R.J.; Riccomagno, E.; Wynn, H.P., Experimental design and observation for large systems, J. roy. statist. soc. ser. B, 58, 77-94, (1996) · Zbl 0850.62627 |

[2] | Chen, J.; Lin, D.K.J., On the identity relationship of 2^{k−p} designs, J. statist. plann. inference, 28, 95-98, (1991) |

[3] | J. B, Clark, and, A. M. Dean, Equivalence of fractional factorial designs, Statist. Sinica, in press. · Zbl 0980.62058 |

[4] | Cohn, J.H.E., On the number of d-optimal designs, J. combin. theory, 66, 214-225, (1994) · Zbl 0802.05018 |

[5] | Draper, N.R.; Mitchell, T.J., Construction of the set of 256-run designs of resolution ⩾5 and set of even 512-run designs of resolution ⩾6 with special reference to the unique saturated designs, Ann. math. statist., 39, 246-255, (1968) · Zbl 0155.27002 |

[6] | Draper, N.R.; Mitchell, T.J., Construction of a set of 512-run designs of resolution ⩾5 and a set of even 1024-run designs of resolution ⩾6, Ann. math. statist., 41, 876-887, (1970) · Zbl 0227.62046 |

[7] | Fang, K.T.; Wang, Y., Number-theoretic methods in statistics, (1994), Chapman & Hall London |

[8] | Fang, K.T.; Winker, D.K.J.; Zhang, P., Uniform design: theory and applications, Technometrics, 42, 237-248, (2000) |

[9] | Fang, K.T.; Mukerjee, R., Connection between uniformity and aberration in regular fractions of two-level factorials, Biometrika, 87, 193-198, (2000) · Zbl 0974.62059 |

[10] | Fries, A.; Hunter, W.G., Minimum aberration 2^{k−p} designs, Technometrics, 22, 601-608, (1980) · Zbl 0453.62063 |

[11] | Hickernell, F.J., A generalized discrepency and quadrature error bound, Math. comp., 67, 299-322, (1998) · Zbl 0889.41025 |

[12] | Hall, M., Hadamard matrices of order 16, Research summary, 1, (1961), Jet Propulsion Lab Pasadena, p. 21-26 |

[13] | Lin, D.K.J.; Draper, N.R., Projection properties of plackett and burman designs, Technometrics, 34, 423-428, (1992) |

[14] | C. X. Ma, K. T. Fang, and, D. K. J. Lin, A note on uniformity and orthogonality, submitted for publication. |

[15] | Masuyama, M., On difference sets for constructing orthogonal arrays of index two and strength two, Rep. statist. appl. res. un. Japan. sci. engrs., 5, 27-34, (1957) |

[16] | Plackett, R.L.; Burman, J.P., The design of optimum multifactorial experiments, Biometrika, 34, 305-325, (1946) · Zbl 0063.06274 |

[17] | Roman, S., Coding and information theory, (1992), Springer-Verlag New York · Zbl 0752.94001 |

This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.