Metamodels for Computer-based Engineering Design: Survey and recommendations

By Timothy W. Simpson, Jesse D. Peplinski, Patrick N. Koch, Janet K. Allen. 2001

2,003 citations recorded by the source corpus.

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

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

Hyper-Parameter Optimization: A Review of Algorithms and Applications (2020, background): “The accuracy of a surrogate model is influenced by sampling points available for model construction and the accuracy of sampling data (Simpson et al., 2001).”
Use of Kriging Models to Approximate Deterministic Computer Models (2005, background): “It is also a good choice for deterministic computer models since it interpolates the observed or known data points [4,5].”
Approximation methods in multidisciplinary analysis and optimization: a panel discussion (2004, background): “polynomial response surfaces, kriging models, radial basis functions, neural networks, multivariate adaptive regression splines), and recent reviews and comparisons of many of these approximation model types can be found in ( Simpson et al. 2001b, c; Jin et al. 2001; Sobieszczanski-Sobieski and Haft”
Tolerance analysis of systems in motion taking into account interactions between deviations (2013, methodology): “’’(34) Moreover, Simpson et al.(6) recommend ANNs for ‘‘repeated, nonlinear and deterministic applications,’’ even as statistical tolerance analyses take into account time-dependent systematic operation-depending deviations and the corresponding interactions toward these deviations.”
A Provably Convergent Multifidelity Optimization Algorithm not Requiring High-Fidelity Derivatives (2010, background): “Cart3D has an adjoint-based mesh refinement, which ensures the error caused by the discretization is less than a tolerance.”
The future of risk assessment (2018, methodology): “This is why simulation has long been advocated as a way to explore and understand system behavior for knowledge retrieval [171,175], and has been used for safety assessment since the 1970s–80s.”
Ensemble of metamodels with optimized weight factors (2008, methodology): “A review of literature provides many examples where the accuracy and efficiency of various metamodels for linear, nonlinear, smooth, and noisy responses have been investigated (Giunta and Watson 1998; Simpson et al. 2001a, b; Jin et al. 2001; Papila et al. 2001; Stander et al. 2004; Clarke et al. ”
Metamodeling in Multidisciplinary Design Optimization: How Far Have We Really Come? (2014, methodology): “[44] 2001 1) Reported the development of several metamodeling techniques and sampling schemes for computer experiments Raised the problem imposed by the high dimensionality and nonlinearity (typically found in computer experiments) 2) Discussed first attempts of metamodeling technique selection Simp”
Digital Twin: Values, Challenges and Enablers From a Modeling Perspective (2019, background): “Besides, another advantage can be viewed in such away that the optimization task may be separated from the generation of data for a ROM via a design of experiments (DOE) approach or other interpolation approaches [305], [306].”
Design and Analysis of Computer Experiments in Multidisciplinary Design Optimization: A Review of How Far We Have Come - Or Not (2008, background): “Hafkta et al. 43 and Simpsonet al. 44 provide extensive reviews of response surface and approximation methods in mechanical and aerospace engineering during this timeframe.”
Reduced-Order Nonlinear Unsteady Aerodynamic Modeling Using a Surrogate-Based Recurrence Framework (2010, methodology): “DOE methods [16,17] are used to select NT initial training cases forwhich to conduct CFDsimulations.”
A non‐stationary covariance‐based Kriging method for metamodelling in engineering design (2007, methodology): “Among the widely used metamodelling techniques [1–8], Kriging is considered powerful and flexible for building surrogate models (or metamodels) of simulated response surfaces with different functional forms [9, 10].”