Case Studies and White Papers

Practical versus Statistical Aspects of Altering Central Composite Designs

Published: July 2003
Author: Mark Anderson

For many central composite designs (CCDs), particularly large ones, the usual alphas put the axial points outside the region of operability. A CCD with an alpha of one, known as a "face centered design" (FCD), avoids this problem by drawing the axial point back onto the face of the hyper cube. However, as the number of FCD factors increase, the correlation among the squared terms in the quadratic in the face-centered cube also increases. For k>5 this causes the variance inflation factors (VIFs) associated with the squared terms to become quite high. As a compromise between FCD and standard CCD, this white paper provides the case for a "practical" alpha of the fourth root of the number of factors (k). For k of 5 or more, this practical alpha balances statistical properties with operational necessities.

Publication: 2003 Joint Statistical Meetings Roundtable Luncheon

How to Use Graphs to Diagnose and Deal with Bad Experimental Data

Published: May 2003
Authors: Mark Anderson, Patrick Whitcomb

This article deals with thorny issues that confront every experimenter how to handle individual results that do not appear to fit with the rest of the data. (A somewhat modified version of this article was published in Quality Engineering. April 2007.)

Publication: ASQ Kansas City Annual Quality Congress

Augmented Ruggedness Testing to Prevent Failures

Published: May 2003
Author: Mark Anderson

Failures in the processing and use of products can often be prevented by applying a form of DOE called ruggedness testing. Check out this article to see how it's done for machine-made bread.

Publication: Quality Progress

DOE Software Paints Picture of Powder Coating Defects

Published: April 2003
Authors: Julia O'Neill, Glenn Correll, Val deBrunce, Ralph Pereida

How do you react when a defect arises within one of your top powder coating products that defies conventional problem-solving techniques? You look for answers with DOE.

Publication: Powder Coating Magazine

Published: February 2003
Authors: Mark Anderson, Shari Kraber

This article explains why standard factorial designs (one array) offer a cost-effective alternative to parameter designs (two array) made popular by Taguchi. It then discusses advanced tools for robust design that involve application of response surface methods (RSM) and measurement of propagation of error (POE).

Publication: Paint & Coatings Industry

Published: October 2002
Author: Mark Anderson

Mix ordinary white glue (Elmer's Glue) and a cross-linking agent: borax (20 MULE TEAM brand from your local grocery store). Eureka! You've made play putty. To make things more interesting, add laundry starch. (STA-FLO concentrated liquid) to the mixture. See how well you can do with this home-made material in comparison to the real thing sold commercially as a toy: Silly Putty.

Published: October 2002
Authors: Gary Oehlert, Patrick Whitcomb

This technical paper details a new class of high-resolution (V) minimum-run two-level factorial designs that characterize two-factor interactions with far fewer runs than classical templates. These "MR5" designs serve well as the core for central composite designs that reduce the number of runs needed for response surface methods.

Publication: Fall Technical Conference 2002

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

Published: March 2002
Author: Stephen Lapin

Design of experiments (DOE) was used to examine the effects of adhesive components in electron beam (EB) curable laminating adhesives for flexible packaging. Adhesives Age.

Publication: Adhesives Age

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.