Course 003: Experiments: Planning, Analysis, and Parameter Design Optimization
(9:00 AM - 4:00 PM)

Instructor: C. F. Jeff Wu, Ph.D., University of Michigan

Course Outline: This one-day course will be based on the book by Jeff Wu and Mike Hamada which will be published by John Wiley in March of 2000. The short course notes will be made available to the participants. This book contains many new methods not found in existing textbooks, and covers more than 80 data sets and 200 exercises. The new tools covered include robust parameter design, use of minimum aberration criterion for optimal factor assignment, orthogonal arrays of economic run size, analysis strategies to exploit interactions, experiments for reliability improvement, and analysis of experiments with non-normal responses. The course will cover basic tools and illustrate them with data from real experiments. The course will be divided into four units:
  • Basic principles for design of experiments including two-level full and fractional factorial designs and basic analysis tools. Techniques for resolving ambiguities in aliased effects and the minimum aberration criterion for optimal factor assignment.
  • Choice of optimal blocking schemes for two-level designs. Three-level fractional factorial designs of 27 and 81 runs. Mixed 2- and 4-level designs. A guide to the use of design tables.
  • Nonregular designs including Plackett-Burman designs and mixed-level orthogonal arrays. Their basic statistical properties and practical use. Use of design tables. Analysis of experiments based on nonregular designs, including the exploitation of interactions
  • Robust parameter design for variation reduction. Layout techniques: cross arrays and single arrays. Modeling strategies. Choice of cross and single arrays and use of design tables. Review on signal-to-noise ratio.
The course is targeting industrial or academic statisticians who are interested in learning modern and effective tools in the design and analysis of experiments. Some basic knowledge on regression analysis and analysis of variance are assumed. No prior background in experimental design is required.

About the Instructor: C. F. Jeff Wu is H. C. Carver Professor of Statistics and also Professor of Industrial and Operations Engineering at the University of Michigan, Ann Arbor. He was formerly the GM/NSERC Chair in Quality and Productivity at the University of Waterloo, which was jointly funded by General Motors of Canada and the Natural Sciences and Engineering Research Council of Canada. He taught at the Statistics Department at the University of Wisconsin from 1977-1988 prior to join U. of Waterloo. He is a Fellow of the IMS and the ASA. He received many prestigious awards including COPSS Award in 1987, Wilcoxon Prize in 1990, Brumbaugh Award in 1992, and Jack Youden Prize in 1997. He gave the 1998 P. C. Mahalanobis Memorial Lecturer at the Indian Statistical Institutes. He received his BS in Mathematics from National Taiwan University in 1971 and Ph.D. in Statistics from the University of California, Berkeley in 1976. His research interests include experimental design, quality and reliability improvement, biotech and preclinical applications, robust product/process design, modeling of complex systems, survey sampling and computer-intensive statistical methods. He is the author or co-author of over 100 research papers and a recent book entitled "Experiments: Planning, Analysis, and Parameter Design Optimization"(Wiley, 2000).

Maximum Class Size: 50. Enrollment will be based on first come first serve. Early registration is encouraged.
Fee: Prior to 4/30/00 Regular $180 Student $90; After 4/30/00 Regular $200 Student $100; please make a note in your registration if you would like to purchase the textbook ($89.95) at door. Lunch is included.


Hosted by www.Geocities.ws

1