SPES Highlights Information for JSM 2016 and FTC 2016

Contributed by James Wendelberger, Los Alamos National Laboratory and SPES Chair-Elect; Ananda Sen, University of Michigan and SPES JSM Chair; and Matthew Pratola, The Ohio State University and SPES Education Chair

Making plans to attend JSM in Chicago? Don’t miss the SPES/Q&P mixer Tuesday at the Hilton in Williford A at 5:30 p.m. Also, consider donating door prizes for the SPES JSM mixer. You can register your donation here.

The SPES program for JSM 2016 will feature five invited sessions, two topic-contributed sessions, four contributed sessions, and one contributed poster session. There are also a number of sessions that are jointly sponsored with the Quality and Productivity Section. Following are our invited and topic-contributed sessions:

Invited Sessions

Statistical Aspects of Computer Experiments, organized by Qiong Zhang of Virginia Commonwealth University

Optimal Experimental Design for Physical Models, organized by Antony Overstall of the University of Glasgow

Online Experimentation: What Is It, Why Use It, and How to Do It Well?, organized by Xinwei Ding of Virginia Tech

Synthesis of Information from Longitudinal Trajectories and Failure Data for Reliability Prediction, organized by Sanjib Basu of Northern Illinois University

Pragmatic Computer Model Calibration in the Modern Big-Simulation/Big Data World, organized by Robert B. Gramacy of The University of Chicago

Topic-Contributed Sessions

Statistical Methods for Remote Sensing, organized by Amy Braverman of Jet Propulsion Laboratory

Estimating the Properties of Physical Time Series by Leveraging the Power of Spectral Analysis, organized by Aaron Springford of Queen’s University at Kingston

SPES to Sponsor FTC 2016 Short Course

SPES will sponsor a short course at the Fall Technical Conference in Minneapolis, Minnesota, in October. The course is titled “Methods for Designing and Analyzing Mixture Experiments” and will be presented by Greg Piepel of the Pacific Northwest National Laboratory.

Course Abstract

Mixture experiments involve changing the proportions of the components of a mixture that make up a product and then observing the resulting changes in the product’s characteristics. The proportions of the components in the mix cannot be varied independently (as in factorial experiments) because they must sum to 1.0 for each run in the experiment. Mixture experiments are useful in many product development areas, including foods and drinks, plastics, alloys, ceramics and glass, gasoline blending, fertilizers, textile fibers, concrete, and drugs.

This short course will provide an overview of various approaches and methods used in designing mixture experiments and analyzing the resulting data. Designs for simplex-shaped and irregular-shaped regions (the latter resulting from placing additional constraints on the component proportions), various types of mixture models that can be fitted to mixture data, graphical techniques for interpreting component effects, and graphical and analytic methods for developing mixtures with optimum properties will be covered. Including process variables and/or a total amount variable in mixture experiments will be discussed. Numerous examples will be used to illustrate these topics.

The course is designed for anyone (statistician or nonstatistician) wanting to know about statistical methods for designing mixture experiments and analyzing the resulting data. Prerequisites are an understanding of elementary statistics concepts and previous exposure to experimental design and least squares regression.

Presenter Biography

Greg Piepel is a laboratory fellow in the Applied Statistics and Computational Analysis group at Pacific Northwest National Laboratory. He works as an applied statistician on multidisciplinary research projects in the physical and engineering sciences and has applied mixture experiment techniques to a wide variety of problems (e.g., glass, ceramics, cement, stainless steel, aluminum production, drugs) over the past 37 years. He has also been an active researcher in the mixture experiment field, with more than 150 publications and technical reports in the areas of mixture experiments, experimental design, and others. He is the developer since 1989 of MIXSOFT, a software toolkit for the design and analysis of mixture experiments. Finally, Piepel is a fellow of the ASA and American Society for Quality (ASQ) and has held several positions with the Physical and Engineering Sciences Section and ASQ.