OVERVIEW OF MODERN DESIGN OF EXPERIMENTS METHODS FOR COMPUTATIONAL SIMULATIONS

By Anthony A. Giunta, Steven F. Wojtkiewicz, Michael S. Eldred. 2003

493 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/4a166199b0df32372f4dd808293796f4892696ab

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

An algorithm for fast optimal Latin hypercube design of experiments (2010, background): “It is well known among designers that the quality of fit, which often defines the performance of the surrogate model during optimization and design space exploration, strongly depends on the DOE (point location and density) [13, 14].”
Metamodeling in Multidisciplinary Design Optimization: How Far Have We Really Come? (2014, background): “Also, our focus is primarily on the surrogate modeling techniques themselves, not the experimental designs used to generate sample data; interested readers are referred to recent overviews and texts on the topic [11,12].”
Incremental modeling of a new high-order polynomial surrogate model (2016, methodology): “Traditionally, the DOE can be categorised as Factor Design (FD) [35], Central Composite Design (CCD) [35], Pseudo-Monte Carlo Sampling [36] (PMCS, e.g. the Latin Hypercube sampling [37], and Orthogonal Sampling [38]) and Quasi-Monte Carlo Sampling (QMCS) [3, 36, 39].”
Monte-Carlo based uncertainty analysis: Sampling efficiency and sampling convergence (2013, background): “Extensive overviews can be found in [50] [51] [52] [53] [54], two popular examples are the 'maximin' and 'uniform' Latin hypercube designs [50] [51].”
Metamodel based high-fidelity stochastic analysis of composite laminates: A concise review with critical comparative assessment (2017, methodology): “There exists both several sampling techniques [9, 117-119] as well as metamodel formation methods [120] as discussed in section 1.”
Design of computer experiments: A review (2017, methodology): “Note that the term orthogonal array is not related to the notion of orthogonality in linear algebra [69].”
Iterative Response Surface Based Optimization Scheme for Transonic Airfoil Design (2007, methodology): “More recent research in DoE methods applied in general to computer experiments [22] suggests that the so-called space filling design of experiments [Latin hypercubes (LHs), Monte Carlo sampling, orthogonal array sampling, Auzde and Englais technique [23] ] are more suitable for the problems encounte”
Design and Analysis of Computer Experiments in Multidisciplinary Design Optimization: A Review of How Far We Have Come - Or Not (2008, background): “50,51,52 Other researchers also started investigating the use of gradient information to facilitate metamodel construction.”
Kriging metamodeling in simulation: A review (2009, background): “, maximin designs and orthogonal arrays) are discussed in [26], including many websites and references; also see [5,17,19,31,48]; a Bayesian view is presented in [6].”
Using Cross Validation to Design Conservative Surrogates (2010, methodology): “When the response y x is expensive to evaluate, we approximate it by a less expensive model ŷ x (surrogate model), based on 1) assumptions on the nature of y x and 2) on the observed values of y x at a set of points, called the experimental design [34,35] (also known as design of experiment).”
An interval uncertain optimization method for vehicle suspensions using Chebyshev metamodels (2014, methodology): “One of the traditional sampling methods is the Design of Experiments (DOEs), which includes a family of methods, such as the Factorial Design (FD) [3], Central Composite Design (CCD) [10], Pseudo-Monte Carlo Sampling (PMCS) [11], and Quasi-Monte Carlo Sampling (QMCS) [11–13].”
Pitfalls of using a single criterion for selecting experimental designs (2008, background): “There have been some studies on the influence of different experimental designs (EDs) on predictions [1–4].”