Probability-Based Latin Hypercube Designs for Irregular Regions

Abstract

Probability-Based Latin Hypercube Designs for Irregular Regions Ying Hung1 1Department of Statistics, Rutgers University, USA ?:? Existing space-filling designs are based on the assumption that the experimental region is rectangular, while in practice this assumption can be violated. A class of probability-based Latin hypercube designs and associated adaptive designs is proposed to accommodate a specific type of irregular regions in a data center thermal management study. The proposed method is applied to obtain an optimal sensor placement plan to monitor and study the thermal distribution in a data center.