Improvement on the Adaptive Response Surface Method for High-Dimensional Computation-Intensive Design Problems

Abstract

Abstract This paper addresses the difficulty of the previously developed Adaptive Response Surface Method (ARSM) for high-dimensional design problems. The ARSM was developed to search for the global design optimum for computation-intensive design problems. This method utilized the Central Composite Designs (CCD), which resulted in an exponentially increasing number of required design experiments. In addition, the ARSM generates a complete new set of CCDs in a gradually reduced design space. These two factors greatly undermine the efficiency of the ARSM. In this work, the Latin Hypercube Designs (LHD) were utilized to generate saturated design experiments. Because of the use of Latin Hypercube Designs, the historical design experiments can be inherited in later iterations. The improved ARSM has been tested using a group of standard testing problems and then applied to an engineering design. In both testing and design application, significant efficiency improvement of the ARSM was observed. The ARSM at the current stage demonstrated strong potential to be an efficient global optimization tool for computation-intensive design problems.