Dual-objective optimal mixture designs

Chongqi Zhang*, Wengkee Wong, Heng PENG

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

Mixture experiments are widely used in many industries and particularly in the manufacture of consumer products. Almost all work to date assumes a single study objective, which is unrealistic. Researchers may want to estimate model parameters and make predictions or extrapolations at the same time. We discuss design issues for determining the optimal proportions of the mixture components when there are two or more objectives in the study and there is a large sample size. We present a general methodology for constructing two types of dual-objective optimal design for mixture experiments and discuss the general applicability of the design strategy to more complicated types of mixture design problems, including mixture experiments.

Original languageEnglish
Pages (from-to)211-222
Number of pages12
JournalAustralian and New Zealand Journal of Statistics
Volume54
Issue number2
DOIs
Publication statusPublished - Jun 2012

Scopus Subject Areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

User-Defined Keywords

  • A, D and I -optimality
  • Approximate design
  • Mixture experiment
  • Multiple-objective optimal design
  • Optimal design

Fingerprint

Dive into the research topics of 'Dual-objective optimal mixture designs'. Together they form a unique fingerprint.

Cite this