Design and Analysis of Deterministic Computer Experiments for Optimization

• Activity Number: 242
• Type: Topic Contributed
• Date/Time: Tuesday, August 13, 2002 : 2:00 PM to 3:50 PM
• Sponsor: Section on Physical & Engineering Sciences*
• Author(s): Martin Meckesheimer*+ and Andrew Booker
• Affiliation(s): The Boeing Company/Phantom Works and The Boeing Company
• Keywords: Optimization ; Computer Experiments ; Metamodeling ; DACE ; Validation

Abstract

In many scientific and engineering domains it is common to analyze and simulate complex physical systems using mathematical models. Although computing resources continue to increase in power and speed, discipline-specific computer simulation modules continue to grow in complexity and remain computationally expensive, limiting their use in design optimization. In order to overcome the computational burden, researchers have developed various approximation strategies as inexpensive metamodels of the discipline-specific (deterministic) simulation models. The analysis of complex engineering systems with metamodels simplifies optimization and/or examination of the system performance over the design space and permits rapid evaluation of alternative designs.

In practical applications, however, the use of approximations introduces additional concerns and requires attention to address issues such as inexpensive error assessment of the metamodel, building metamodels for high-dimensional problems for use in optimization. This paper presents an overview of the practical needs that will direct future research efforts in the area of metamodeling.