Exclusive Content & Downloads from ASQ

Illustration of Confounding in Plackett-Burman Designs

Summary: [This abstract is based on the author's abstract.]

Plackett-Burman designs represent powerful tools for process improvement. Appropriate use and interpretation of the designs requires an understanding of their confounding characteristics. The confounding principles of Plackett-Burman designs are illustrated, and the differences in geometric and nongeometric types are discussed. The impact of confounding on conclusions drawn from an experiment is shown to provide a better understanding of how to interpret experimental results. An understanding of basic design analysis is assumed. Plackett-Burman designs require "n" trials and to study "n-1" factors, where "n" is divisible by 4. These highly fractionated designs confound perfectly a few two-factor interactions with main effects in the geometric case. They confound partially many two-factor interactions with main effects in the nongeometric case. The effects of the interactions are added to the associated main effect. Models simulate these results and show the impact. The value of reflection as a protection against wrong conclusions becomes apparent. Assumptions of "effect heredity" and "effect sparsity" are discussed. The role of these assumptions in providing meaningful results in the presence of confounding is examined.

Anyone with a subscription, including Site and Enterprise members, can access this article.

Other Ways to Access content:

Join ASQ

Join ASQ as a Full member. Enjoy all the ASQ member benefits including access to many online articles.

  • Topics: Process Management
  • Keywords: Screening,Fractional factorial design,Plackett and Burman designs,Interactions
  • Author: Barrentine, Larry B.
  • Journal: Quality Engineering