Case Studies and White Papers


Published: October 2002
Authors: Mark Anderson, Patrick Whitcomb

This article offers a simple case study that illustrates how to put rubber or plastics formulations to the test by using powerful statistical methods for mixture design and analysis. Rubber & Plastics News.

Publication: Rubber & Plastics News

Screening Ingredients Most Efficiently with Two-Level Design of Experiments (DOE)

Published: February 2002
Author: Mark Anderson

A DOE on machine-made bread shows how clever application of statistical methods quickly screens alternative ingredients to see which, if any, impair the desired reaction.

Published: February 2002
Author: Mark Anderson

This kitchen experiment on a bread-baking machine illustrates the power of multifactor testing for unveiling breakthrough interactions. The surprising results from the original two-level fractional factorial were confirmed by an innovative follow-up experiment called a "semi-foldover".

Publication: Today's Chemist at Work

Cost-Effective and Information-Efficient Robust Design for Optimizing Processes and Accomplishing Six Sigma Objectives

Published: January 2002
Authors: Mark Anderson, Shari Kraber

Standard factorial designs (one array) offer a cost-effective and information-efficient robust design alternative to parameter designs (two-array) made popular by Taguchi. This paper compares these two methods (one-array versus two-array) in depth via an industrial case study. It then discusses advanced tools for robust design that involve application of response surface methods (RSM) and measurement of propagation of error (POE).

Publication: Society of Manufacturing Engineers

How to Save Runs, Yet Reveal Breakthrough Interactions, By Doing Only a Semifoldover on Medium-Resolution Screening Designs

Published: May 2001
Authors: Mark Anderson, Patrick Whitcomb

Via case studies, this paper reviews the strategy of foldover on low-resolution (III) two-level fractional factorials and demonstrates how to reduce experimental runs by making use of semifoldover methods to augment medium-resolution (IV) designs.

Publication: ASQC 55th Annual Quality Congress Proceedings

Achieving Six Sigma Objectives for Variability Reduction in Formulation and Processing

Published: January 2001
Authors: Mark Anderson, Patrick Whitcomb

Apply powerful design of experiments (DOE) tools to make your system more robust to variations in component levels and processing factors.

Design Experiments that Combine Mixture Components with Process Factors

Published: December 2000
Authors: Mark Anderson, Patrick Whitcomb

This article shows how to do a comprehensive experiment that combines mixture components with process factors in one crossed design, thus revealing interactions that would remain hidden by not combining all the variables in one study.

Publication: Chemical Engineering Progress

Practical Aids for Teaching Experimental Designs

Published: February 2000
Authors: Madhuri Mulekar, Mark Anderson, D.W. McCarmik, Pat Spagon

Design of Experiments (DOE) is an essential tool for product and process improvement. Good software now makes the set up for design and analysis of experiments very easy, but many engineers and/or non-statisticians feel intimidated by statistical outputs. For that reason, non-statisticians need training in proper designing and conducting of experiments. Ideally the DOE training is best when provided on a just-in-time basis - prior to actually doing an experiment. However, an in-class experiment is a reasonable substitute for real-life experiments. It is important for technicians to gain an understanding of designed experiments so that mistakes made in conducting the experiment are reduced and the data is collected more correctly and accurately from the experiments. Managers read and possibly edit reports, and make decisions based on the experiments run by technicians and engineers. An understanding of design and analysis techniques helps them identify any problems with the experiment or the results. At SEMATECH, there are two general audiences. The first is comprised of project engineers. They are responsible for designing and conducting experiments to determine if the project they work on meets its goals. Most of them take an eight day long course on designed experimentation that covers many basic concepts. The second group is comprised of project managers and technicians. This group does not have as great a need for the details of experimental design as the first group. Project managers manage the project engineers, and the technicians carry out the experimentation under the direction of the project engineers. The technicians also assist the project engineers in running the equipment on which the experiments are conducted. This helps ensure that experimental runs and data collection proceed smoothly. The project managers oversee all aspects of the experiment and must understand the outcomes of the experiments too. So the managers must provide adequate resources for the experiment, must know why certain designs are used, and be able to interpret and critique the analysis intelligently. The emphasis of the exercise is to motivate, illustrate and provide hands on experience with the methodologies and analysis techniques discussed in class.

How to Design and Analyze Mixture Designs that Include Process Factors and/or Categorical Variables

Published: January 2000
Authors: Mark Anderson, Patrick Whitcomb

The latest versions of dedicated DOE software exhibit more versatility than ever before to create optimal designs that handle any combination of mixture components, processing factors (such as time or temperature) and categorical variables (such as supplier and material type). These computer programs easily manipulate almost any number of responses in powerful optimization routine that reveal "sweet spots" - the operating windows that meet all specifications at minimal cost. In this paper, we review the basic principles of mixture design. Then we apply state-of-the-art tools for optimal design to the formulation of a coating.

Publication: Industrial Statistics in Action 2000

Eliminate Nuisance Variability with a Latin Square Design

Published: January 2000
Authors: Douglas Montgomery, Mark Anderson

Case study from Douglas Montgomery's <em>Design and Analysis of Experiments, 4th Edition</em>, John Wiley, Example 5-4 on page 196.