Accession Number : ADA329501

Title :   Numerical Optimization Using Computer Experiments

Corporate Author : INSTITUTE FOR COMPUTER APPLICATIONS IN SCIENCE AND ENGINEERING HAMPTON VA

Personal Author(s) : Trosset, Michael W. ; Torczon, Virginia

PDF Url : ADA329501

Report Date : AUG 1997

Pagination or Media Count : 16

Abstract : Engineering design optimization often gives rise to problems in which expensive objective functions are minimized by derivative free methods. We propose a method for solving such problems that synthesizes ideas from the numerical optimization and computer experiment literatures. Our approach relies on kriging known function values to construct a sequence of surrogate models of the objective function that are used to guide a grid search for a minimizer. Results from numerical experiments on a standard test problem are presented.

Descriptors :   *COMPUTERIZED SIMULATION, *COMPUTER AIDED DESIGN, *EXPERIMENTAL DESIGN, ALGORITHMS, SOFTWARE ENGINEERING, OPTIMIZATION, STOCHASTIC PROCESSES, MATRICES(MATHEMATICS), APPROXIMATION(MATHEMATICS), SUBROUTINES.

Subject Categories : Statistics and Probability
      Computer Programming and Software

Distribution Statement : APPROVED FOR PUBLIC RELEASE