A Graphical Tool for Detection of Outliers in Completely Randomized, Unreplicated 2^k and 2^k-P Factorials

ABSTRACT

With the increased awareness of statistical methods in industry today, many non-statisticians are implementing statistical studies and conducting statistically designed experiments (DOEs). With this increased use of DOEs by non-statisticians in applied settings, there is a need for more graphical methodologies to support both analysis and interpretations of DOE results. In particular, there is a critical need for user-friendly means to investigate outlier effects of noise and active background variables in unreplicated DOEs. This article presents a profoundly simple, yet effective, methodology to identify outliers in unreplicated 2 ^k and 2 ^k-p factorial designs that integrates well-established, confirmatory statistical techniques with a simple graphical, exploratory tool.

KEYWORDS:

• design of experiments
• DOE
• fractional factorials
• graphical analysis
• outliers
• unreplicated experimentation

Additional information

Notes on contributors

Doug Sanders

Doug Sanders, Ph.D. is a faculty member of the Center for Executive Education and President of Six Sigma Associates of Tennessee. He has guided the global transformational efforts of industrial and commercial organizations worldwide. Sanders received his M.S. in Statistics and his Ph.D. in Management Science from the University of Tennessee.

Cheryl Hild

Dr. Cheryl Hild is Director of Quality at Aegis Sciences Corporation, Nashville Tennessee. She serves as an adjunct faculty member for the University of Tennessee's Center for Executive Education. She served over nine years as a senior associate with Six Sigma Associates. Dr. Hild has authored numerous peer-reviewed articles and is co-author of the book, The Power of Statistical Thinking. She received Ph.D. in Management Science and Statistics from the University of Tennessee.