Announcing a
1-Day Short Course on
Methods for Designing and Analyzing
Mixture Experiments
In conjunction with the 2008 Fall
Technical Conference in Phoenix, AZ
Sponsored by the Statistics Division of
the American Society for Quality
October 8 (Wednesday), 2008
John A. Cornell, Emeritus Professor, Department of
Statistics,
Greg F. Piepel, Statistics and Sensor Analytics,
Pacific Northwest National Laboratory,
greg.piepel@pnl.gov
or mixsoft@aol.com
People perform mixture experiments
daily. Imagine adding sugar and/or a
sweetener to coffee or tea to improve the flavor, or the mixing oil and vinegar
and some spices to create and flavor a salad dressing, or the addition of 93
octane fuel (premium unleaded) in the fuel tank of the family car which
presently contains some 87 or 89 octane fuel in hopes of getting better mileage
or performance. In other words, adding
and/or blending ingredients in an effort to obtain a more desirable end product
is something all of us do in our everyday
activities. These actions are known as
performing mixture experiments
and are of great use in developing or improving many commercial and research
products.
Performing
mixture experiments consists of varying the proportions (or percentages) of the
individual ingredients in an attempt to see if measured quality characteristics
or properties of the end product change from one blend to the next. For example, a property of interest of
stainless steel (which is a mixture of iron, nickel, copper, and chromium) is
its tensile strength. Changing the
relative proportions of iron, nickel, and copper, while holding the proportion
of chromium fixed, will certainly affect the tensile strength. Changing the percentages of orange,
pineapple, and grapefruit juices, respectively, from 30%-35%-35% to 40%-50%-10%
would certainly change the flavor of the three-juice beverage.
This course will begin by addressing the
questions, “What are mixture experiments?” and “How do mixture experiments
differ from ordinary experiments such as factorial experiments?” You will learn how to construct mixture
designs and how to develop and fit mixture models to measure the blending
properties of the mixture components as well as to measure product
performance. Other topics to be covered
are screening mixture components, including process variables in mixture
experiments, methods for optimizing formulations, and software that is
available for designing and analyzing data from mixture experiments. Several worked examples from the initial
design to inferences made from the analysis of data are taken from real
experiments. Copies of course notes
matching overhead materials will be provided to all attendees.
The course presenters have a combined 70+
years working with mixture experiments. They welcome questions about the course
and are willing to address questions you may have about your specific mixture
experiences. Please address such
questions by emailing John or Greg at the email addresses below their names at
the top of the sheet. For additional
information on topics to be covered in the course, longer versions of the
course, or about either of the two presenters, please go to http://members.aol.com/mixsoft/mixsc.htm. For more information about the 2008 Fall
Technical Conference, see FTC 2008.
Number of Visits to This Page
Return to Table of Contents