Maximum Projection Designs for Computer Experiments

• Activity Number: 501
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Physical and Engineering Sciences
• Author(s): Shan Ba* and V. Roshan Joseph and Evren Gul
• Companies: Procter & Gamble Company and Georgia Institute of Technology and Georgia Institute of Technology
• Keywords: Experimental design ; Gaussian process ; Latin hypercube design ; Screening design ; Space-filling design

Abstract

Space-filling properties are important in designing computer experiments. The traditional maximin and minimax distance designs only consider space-filling in the full dimensional space. This can result in poor projections on lower dimensional spaces, which are undesirable when only a few factors are active. Restricting maximin distance design to the class of Latin hypercubes can improve one-dimensional projections, but cannot guarantee good space-filling properties in the other subspaces. In this work, we propose designs that maximize certain space-filling properties on projections to all subsets of factors. We call our designs maximum projection designs. Our design criterion can be computed at a cost no more than a design criterion that ignores projection properties.