Using Multiple Objectives to Select an Optimal Split-Plot Design for a Mixture-Process Experiment

• Activity Number: 104
• Type: Invited
• Date/Time: Monday, August 10, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Physical and Engineering Sciences
• Author(s): Christine Anderson-Cook* and Lu Lu and Timothy Robinson
• Companies: Los Alamos National Laboratory and University of South Florida and University of Wyoming
• Keywords: Pareto Front ; Restricted Randomization ; Trade-offs

Abstract

With increasingly constrained budgets, it is now becoming more desirable to get more information from each experiment and to have an intentional strategy for selecting a split-plot design (SPD) that balance multiple competing objectives. We consider a case study of a mixture-process experiment that seeks an SPD with a good balance of precise model coefficient estimates and low experimental cost, which is a function of time and financial cost. The D-efficiency is a function of the whole plot-to-subplot error variance ratio, a quantity that is typically not known a priori when the choice of a design must be made. Substantially different patterns of design performance robustness to the uncertainty of the specified variance ratio is demonstrated with the mixture-process design compared to non-mixture experiments.