Advanced search
2 files | 14.11 MB Add to list

Sequential sensitivity analysis of expensive black-box simulators with metamodelling

Tom Van Steenkiste (UGent) , Joachim van der Herten (UGent) , Ivo Couckuyt (UGent) and Tom Dhaene (UGent)
Author
Organization
Abstract
Simulators are an essential tool in the design and analysis of complex systems. For many tasks in engineering, highly accurate simulators are used instead of expensive real-life experiments. As accuracy increases, these simulators are becoming computationally more expensive to evaluate. Sensitivity analysis provides information on the importance of the inputs with regard to the outputs. This does not only provide critical information about the workings of the system but also allows to discard inputs with low impact on the outputs. However, sensitivity analysis is an expensive process in terms of number of evaluations. Hence, metamodelling techniques are used to reduce the computational burden. In this paper we present, discuss and evaluate a novel algorithm for sequential variance-based and derivative-based sensitivity analysis of expensive black-box simulators using metamodelling. Two new stopping criteria are proposed on top of the traditional model error based stopping criteria. Extensive testing on benchmark problems and engineering use cases shows the sensitivity indices can be efficiently and accurately computed with a limited number of simulator evaluations. (C) 2018 Elsevier Inc. All rights reserved.
Keywords
COMPUTER EXPERIMENTS, GAUSSIAN-PROCESSES, DESIGN STRATEGY, INDEXES, TOOLBOX, MODEL, OPTIMIZATION, CODES, Sensitivity analysis, Metamodelling, Simulations, Design of experiments, Engineering design

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 3.77 MB
  • 7208 i.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 10.34 MB

Citation

Please use this url to cite or link to this publication:

MLA
Van Steenkiste, Tom, Joachim van der Herten, Ivo Couckuyt, et al. “Sequential Sensitivity Analysis of Expensive Black-box Simulators with Metamodelling.” APPLIED MATHEMATICAL MODELLING 61 (2018): 668–681. Print.
APA
Van Steenkiste, T., van der Herten, J., Couckuyt, I., & Dhaene, T. (2018). Sequential sensitivity analysis of expensive black-box simulators with metamodelling. APPLIED MATHEMATICAL MODELLING, 61, 668–681.
Chicago author-date
Van Steenkiste, Tom, Joachim van der Herten, Ivo Couckuyt, and Tom Dhaene. 2018. “Sequential Sensitivity Analysis of Expensive Black-box Simulators with Metamodelling.” Applied Mathematical Modelling 61: 668–681.
Chicago author-date (all authors)
Van Steenkiste, Tom, Joachim van der Herten, Ivo Couckuyt, and Tom Dhaene. 2018. “Sequential Sensitivity Analysis of Expensive Black-box Simulators with Metamodelling.” Applied Mathematical Modelling 61: 668–681.
Vancouver
1.
Van Steenkiste T, van der Herten J, Couckuyt I, Dhaene T. Sequential sensitivity analysis of expensive black-box simulators with metamodelling. APPLIED MATHEMATICAL MODELLING. New york: Elsevier Science Inc; 2018;61:668–81.
IEEE
[1]
T. Van Steenkiste, J. van der Herten, I. Couckuyt, and T. Dhaene, “Sequential sensitivity analysis of expensive black-box simulators with metamodelling,” APPLIED MATHEMATICAL MODELLING, vol. 61, pp. 668–681, 2018.
@article{8574361,
  abstract     = {Simulators are an essential tool in the design and analysis of complex systems. For many tasks in engineering, highly accurate simulators are used instead of expensive real-life experiments. As accuracy increases, these simulators are becoming computationally more expensive to evaluate. Sensitivity analysis provides information on the importance of the inputs with regard to the outputs. This does not only provide critical information about the workings of the system but also allows to discard inputs with low impact on the outputs. However, sensitivity analysis is an expensive process in terms of number of evaluations. Hence, metamodelling techniques are used to reduce the computational burden. In this paper we present, discuss and evaluate a novel algorithm for sequential variance-based and derivative-based sensitivity analysis of expensive black-box simulators using metamodelling. Two new stopping criteria are proposed on top of the traditional model error based stopping criteria. Extensive testing on benchmark problems and engineering use cases shows the sensitivity indices can be efficiently and accurately computed with a limited number of simulator evaluations. (C) 2018 Elsevier Inc. All rights reserved.},
  author       = {Van Steenkiste, Tom and van der Herten, Joachim and Couckuyt, Ivo and Dhaene, Tom},
  issn         = {0307-904X},
  journal      = {APPLIED MATHEMATICAL MODELLING},
  keywords     = {COMPUTER EXPERIMENTS,GAUSSIAN-PROCESSES,DESIGN STRATEGY,INDEXES,TOOLBOX,MODEL,OPTIMIZATION,CODES,Sensitivity analysis,Metamodelling,Simulations,Design of experiments,Engineering design},
  language     = {eng},
  pages        = {668--681},
  publisher    = {Elsevier Science Inc},
  title        = {Sequential sensitivity analysis of expensive black-box simulators with metamodelling},
  url          = {http://dx.doi.org/10.1016/j.apm.2018.05.023},
  volume       = {61},
  year         = {2018},
}

Altmetric
View in Altmetric
Web of Science
Times cited: