Abstract #301781

This is the preliminary program for the 2003 Joint Statistical Meetings in San Francisco, California. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 2-5, 2003); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2003 Program page
JSM 2003 Abstract #301781
Activity Number: 23
Type: Contributed
Date/Time: Sunday, August 3, 2003 : 2:00 PM to 3:50 PM
Sponsor: Section on Physical and Engineering Sciences
Abstract - #301781
Title: A Two-Stage Layered Mixture Experiment Design for a Nuclear Waste Glass Application (Part 1)
Author(s): Gregory F. Piepel*+ and Scott K. Cooley and Hao Gan and Wing Kot and Ian Pegg
Companies: Battelle, PNNL and Battelle-Pacific Northwest Division and Catholic University of America and Catholic University of America and Catholic University of America
Address: 649 Cherrywood Loop, Richland, WA, 99352-1807,
Keywords: layered design ; mixture experiment ; optimal experimental design ; nuclear waste glass
Abstract:

A layered experimental design involving mixture variables was generated to support developing waste glass property-composition models. The design was generated in two stages, each having unique characteristics. Each stage used a layered design having an outer layer, an inner layer, a center point, and some replicates. The layers were defined by single- and multi-variable constraints. The first stage involved 15 glass components treated as mixture variables. For each layer, vertices were generated and optimal design software was used to select alternative subsets of vertices and calculate design optimality measures. Two partial quadratic mixture models, containing 25 terms for the outer layer and 30 terms for the inner layer, were the basis for the optimal design calculations. Distributions of predicted glass property values were plotted and evaluated for the alternative subsets of vertices. Based on the optimality measures and the predicted property distributions, a 'best' subset of vertices was selected for each layer to form a layered design for the first stage. The design for the second stage was selected to augment the first-stage design, and is the subject of Part 2.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2003 program

JSM 2003 For information, contact meetings@amstat.org or phone (703) 684-1221. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2003