ASQ

Exclusive Content & Downloads from ASQ

Analysis of a Difficult Factorial Designed Experiment from Polyurethane Product Research

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

In the plastic or polymer product field, researchers often investigate processing and additive effects on polymer material properties in order to improve product performance. Polymer performance is extremely variable and difficult to control. To understand how the performance correlates to the basic polymer microstructure, several analytical methods that measure microstructural features are applied. Well-planned experiments are necessary in this approach. Variability can make it difficult to determine which factors are really significant even with a designed experiment approach. A case study from research on polyurethane products is presented. The study shows a variety of approaches for checking for outliers and determining reasonable factor effects. The study examines three factors that potentially affect the polyurethane microstructure. A 5 X 4 X 3 factorial designed experiment is presented and demonstrates the difficulty with statistical analysis of factor effects when high variability exists. A few data points are suspected as potential outliers, while two of the factors and their interaction are always found to be significant. The third factor is found significant only after the apparent outliers are eliminated from the analysis. A rank transformation is applied to minimize the impact of outliers and to aid in determining which factors are significant.

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: Design of Experiments
  • Keywords: Case study,Chemical and process industries,Transformation,Experiments,Factorial experiments,Design of experiments (DOE)
  • Author: Heaney, Michael D.; Lidy, Werner A.; Rightor, Edward G.; Barnes, C. Glenn
  • Journal: Quality Engineering