Computation-intensive design problems are becoming increasingly common in manufacturing industries. The computation burden is often caused by expensive analysis and simulation processes in order to reach a comparable level of accuracy as physical testing data. To address such a challenge, approximation or metamodeling techniques are often used. Metamodeling techniques have been developed from many different disciplines including statistics, mathematics, computer science, and various engineering disciplines. These metamodels are initially developed as “surrogates” of the expensive simulation process in order to improve the overall computation efficiency. They are then found to be a valuable tool to support a wide scope of activities in modern engineering design, especially design optimization. This work reviews the state-of-the-art metamodel-based techniques from a practitioner’s perspective according to the role of metamodeling in supporting design optimization, including model approximation, design space exploration, problem formulation, and solving various types of optimization problems. Challenges and future development of metamodeling in support of engineering design is also analyzed and discussed.Copyright © 2006 by ASME

https://www.sfu.ca/~gwa5/index_files/Microsoft%20Word%20-%20ASME-Metamodeling-6-12-06.pdf

Review of Metamodeling Techniques in Support of Engineering Design Optimization

This paper addresses the solution of bound-constrained optimization problems using algorithms that require only the availability of objective function values but no derivative information. We refer to these algorithms as derivative-free algorithms. Fueled by a growing number of applications in science and engineering, the development of derivative-free optimization algorithms has long been studied, and it has found renewed interest in recent time. Along with many derivative-free algorithms, many software implementations have also appeared. The paper presents a review of derivative-free algorithms, followed by a systematic comparison of 22 related implementations using a test set of 502 problems. The test bed includes convex and nonconvex problems, smooth as well as nonsmooth problems. The algorithms were tested under the same conditions and ranked under several criteria, including their ability to find near-global solutions for nonconvex problems, improve a given starting point, and refine a near-optimal solution. A total of 112,448 problem instances were solved. We find that the ability of all these solvers to obtain good solutions diminishes with increasing problem size. For the problems used in this study, TOMLAB/MULTIMIN, TOMLAB/GLCCLUSTER, MCS and TOMLAB/LGO are better, on average, than other derivative-free solvers in terms of solution quality within 2,500 function evaluations. These global solvers outperform local solvers even for convex problems. Finally, TOMLAB/OQNLP, NEWUOA, and TOMLAB/MULTIMIN show superior performance in terms of refining a near-optimal solution.

/pdf/derivative-free-optimization-a-review-of-algorithms-and-2qmveq3j89.pdf

Derivative-free optimization: a review of algorithms and comparison of software implementations

This article reviews Kriging (also called spatial correlation modeling). It presents the basic Kriging assumptions and formulas, contrasting Kriging and classic linear regression metamodels. Furthermore, it extends Kriging to random simulation, and discusses bootstrapping to estimate the variance of the Kriging predictor. Besides classic one-shot statistical designs such as Latin Hypercube Sampling, it reviews sequentialized and customized designs. It ends with topics for future research.

/pdf/kriging-metamodeling-in-simulation-a-review-2hj4mjqeb7.pdf

Kriging Metamodeling in Simulation: A Review

Kriging metamodeling in simulation : A review

The use of kriging models for approximation and metamodel-based design and optimization has been steadily on the rise in the past decade. The widespread usage of kriging models appears to be hampered by (1) the lack of guidance in selecting the appropriate form of the kriging model, (2) computationally efficient algorithms for estimating the model’s parameters, and (3) an effective method to assess the resulting model’s quality. In this paper, we compare (1) Maximum Likelihood Estimation (MLE) and Cross-Validation (CV) parameter estimation methods for selecting a kriging model’s parameters given its form and (2) and an R 2 of prediction and the corrected Akaike Information Criterion for assessing the quality of the created kriging model, permitting the comparison of different forms of a kriging model. These methods are demonstrated with six test problems. Finally, different forms of kriging models are examined to determine if more complex forms are more accurate and easier to fit than simple forms of kriging models for approximating computer models.

http://www2.mae.ufl.edu/haftka/eoed/protected/ASME04.kriging.pdf

Use of Kriging Models to Approximate Deterministic Computer Models

This paper summarizes the discussion at the Approximation Methods Panel that was held at the 9
 th
 AIAA/ISSMO Symposium on Multidisciplinary Analysis & Optimization in Atlanta, GA on September 2–4, 2002. The objective of the panel was to discuss the current state-of-the-art of approximation methods and identify future research directions important to the community. The panel consisted of five representatives from industry and government: (1) Andrew J. Booker from The Boeing Company, (2) Dipankar Ghosh from Vanderplaats Research & Development, (3) Anthony A. Giunta from Sandia National Laboratories, (4) Patrick N. Koch from Engineous Software, Inc., and (5) Ren-Jye Yang from Ford Motor Company. Each panelist was asked to (i) give one or two brief examples of typical uses of approximation methods by his company, (ii) describe the current state-of-the-art of these methods used by his company, (iii) describe the current challenges in the use and adoption of approximation methods within his company, and (iv) identify future research directions in approximation methods. Several common themes arose from the discussion, including differentiating between design of experiments and design and analysis of computer experiments, visualizing experimental results and data from approximation models, capturing uncertainty with approximation methods, and handling problems with large numbers of variables. These are discussed in turn along with the future directions identified by the panelists, which emphasized educating engineers in using approximation methods.

Approximation methods in multidisciplinary analysis and optimization: a panel discussion

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, computer simulation and analysis codes continue to grow in complexity and remain computationally expensive, limiting their use in design and optimization. Consequently, many researchers have developed different metamodeling strategies to create inexpensive approximations of computationally expensive computer simulations. These approximations introduce a new element of uncertainty during design optimization, and there is a need to develop efficient methods to assess metamodel validity. We investigate computationally inexpensive assessment methods for metamodel validation based on leave-k-out cross validation and develop guidelines for selecting k for different types of metamodels. Based on the results from two sets of test problems, k = 1 is recommended for leave-k-out cross validation of low-order polynomial and radial basis function metamodels, whereas k=0.1N or N is recommended for kriging metamodels, where N is the number of sample points used to construct the metamodel.

Computationally Inexpensive Metamodel Assessment Strategies

Experimental design methods can be applied to engineering design activities to understand which variables affect the system under consideration, how these variables affect the system, and how to select variable settings that will give uniformly long life to the system. The objective of this paper is to demonstrate the use of Design and Analysis of Computer Experiment (DACE) methods (Sacks, J. et al., 1989) and design optimization via the Surrogate Management Framework (Booker, A. J. et al., 1999; Audet, C. et al., 2000) on reliability optimization problems. Reliabilities are calculated using the Probabilistic Structural Analysis Method (Palle Thoft–Christensen and Baker, 1982; Achintya Haldar and Sankaran Mahadevan, 2000), a method for estimation of reliabilities and reliability indices for a structural model given probability distributions for design variables and “environmental” variables such as loads. By maximizing reliability, or minimizing the probability of failure, we attempt to achieve a minimum cost design that is affected minimally by the variability in the design variables.

Reliability Based Design Optimization Using Design Explorer

Inmanyscientie candengineeringdomains,itiscommonto analyzeandsimulatecomplexphysicalsystemsusing mathematicalmodels.Althoughcomputingresourcescontinuetoincreaseinpowerandspeed,computersimulation andanalysiscodescontinuetogrowincomplexityandremaincomputationallyexpensive,limitingtheiruseindesign and optimization. Consequently, many researchers have developed different metamodeling strategies to create inexpensive approximations of computationally expensive computer simulations. These approximations introduce a new element of uncertainty during design optimization, and there is a need to develop efe cient methods to assess metamodel validity. We investigate computationally inexpensive assessment methods for metamodel validation based on leave- k-out cross validation and develop guidelines for selecting k for different types of metamodels. Based on the results from two sets of test problems, k= 1 is recommended for leave- k-out cross validation of loworder polynomial and radial basis function metamodels, whereas k=0:1N or N is recommended for kriging metamodels, where N is the number of sample points used to construct the metamodel.

Andrew J. Booker

Papers

Approximation methods in multidisciplinary analysis and optimization: a panel discussion

Computationally Inexpensive Metamodel Assessment Strategies

Reliability Based Design Optimization Using Design Explorer

Computationally inexpensive metamodel assessment strategies