CQAS Seminars

 

Implementation of Six Sigma with Regression Methods

 

Many people working on Six Sigma projects do not realize the additional benefits of utilizing better regression models. Analysis of variance (ANOVA) models used in designed experiments are special cases of regression models. They can also be extended to model covariates, such as time, ambient temperature or other uncontrolled factors which are trying to "sneak" into your experiment. Regression models are the most versatile tools for data analysis.

In this hands-on, two-day seminar, you will first learn how to use regression techniques in a wide variety of contexts, from designed experiments to observational studies, especially those performed in Six Sigma projects. You will also learn about dangers of regression analysis and how to avoid them. You will then practice those skills in problem-solving sessions and learn how to apply them using MINITAB® computer software. In addition, you will learn about statistical resources available on the Internet. No prior knowledge of regression methods, or MINITAB® is required. This course provides 1.4 CEUs.

 

Topics include:

• Simple linear regression
• Multiple linear regression
• Diagnostic tools
• Estimation and prediction using regression
• Polynomial regression
• Transformation of variables
• Dangers of regression analysis and how to avoid them
• Case studies in the use of regression methods

How you will benefit:

• Applying more powerful tools in the "Analyze" and "Improve" phases of your Six Sigma projects
• Extracting the maximum amount of information from your data
• Speeding up your work by an efficient use of modeling tools and easy-to-use statistical software, MINITAB®
• Utilizing statistical resources available on the Internet

 

Cost

 

$695. This price includes a copy of The Manual for Statistical Resources on the Internet, written by the instructor. Continental breakfast and lunch is included each day of the seminar.

 

Meet your instructor

Peter Bajorski is a Ph.D. statistician with 20 years of experience in consulting, research, and teaching. He is currently a faculty member at CQAS. Prior to joining RIT, he held positions at Cornell University, the University of British Columbia, and Simon Fraser University. He was also an Associate Statistician at the Engineering Research and Development Bureau, New York State Department of Transportation.

 

Peter is a Six Sigma Black Belt and is familiar with the Lean Enterprise approach to process improvement. In addition to his expertise in process improvement methods, Peter's statistical focus is in regression techniques, multivariate analysis, design of experiments, and nonparametric methods. He also has done statistical work in reliability, imaging science, transportation, health services, quality assurance and material engineering, civil engineering, and industrial engineering. Peter has authored five short-course workbooks.