This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).
This grant provides funding for developing experimental design theories and techniques for product and process reliability responses that will fill the gap that currently exists in design for reliability. Direct application of traditional design of experiment (DOE) tools on reliability experiments is inappropriate. This project will attack the fundamental problems of reliability experiments, including non-normal response distribution, nonlinear model and parameter dependency problems, which have prevented the efficient use of DOE for product and process reliability improvement. Preliminary studies indicate that the proposed methodology will result in designs superior to those encountered in practice. This project is innovative and transformative in terms of creating and testing a group of new DOE tools that will greatly expand the capability and applicability of statistical experimental design.
If successful, the theoretical results foreseen in the proposal will impact a number of scientific communities, including engineering design, material design, biomedicine and clinical trails. By "designing-in" reliability, companies will be able to reduce the number of experimental runs needed throughout the product design, development, and delivery. A comprehensive website devoted to experimental design and reliability will be developed to disseminate the results of this research. The proposed research activities are a natural venue for training graduate and undergraduate students to develop theoretical models and applications of experimental design and reliability analysis. This project will boost the interdisciplinary research (Industrial Engineering and Mathematical Statistics) at two universities (Arizona State University and Southern Illinois University Edwardsville), and improve both theoretical and application types of education for engineering and mathematical science students.
