Abstract
Graphical presentations have been used by many statisticians and engineers to view the properties of particular two- and three-factor factorial and fractional-factorial designs, and to communicate the nature of a planned experiment to production workers and management. The geometric characteristics of the graphical representation are related to the statistical properties of the design. These geometric characteristics can be examined for irregular designs, when defining relations do not exist. This paper examines design-plots for factorial and fractional-factorial designs involving as many as ten factors; and illustrates that even these higher-order design-plots can be used to display the results of an experiment and to provide useful insight on the nature of the response surface.
Additional information
Notes on contributors
Russell R. Barton
Dr. Barton is an Associate Professor in the Department of Industrial and Manufacturing Engineering. He is a member of ASQ.