1. Proceedings
  2. I3M
  3. 2018
  4. EMSS
  5. 10.46354/i3m.2018.emss.032

An efficient global sensitivity analysis method based on sequential Latin hypercube sampling

  • a Lingyun Lu  ,
  • b Wei Li   ,
  • c Ping Ma  , 
  • d Ming Yang  
  • Control and Simulation Center, Harbin Institute of Technology, Harbin 150080, P.R. China
Cite as
Lu L., Li W., Ma P., Yang M. (2018). An efficient global sensitivity analysis method based on sequential Latin
hypercube sampling. Proceedings of the 30th European Modeling & Simulation Symposium (EMSS 2018), pp. 234-240. DOI: https://doi.org/10.46354/i3m.2018.emss.032

Abstract

Many simulation models involve inputs and parameters, which are not precisely known. Global sensitivity analysis aims to identify these inputs and parameters whose uncertainty has the largest impact on the variability of model output. In this paper, an efficient global sensitivity analysis method based on sequential
Latin hypercube sampling is proposed. Firstly, the basic theory of Sobol’ method is formulized and a generalized estimator of first order sensitivity analysis indices is proposed. Then, a sequential sampling strategy based on extended optimal Latin hypercube sampling is adopted to improve the sampling efficiency of sensitivity analysis. Finally, the proposed method is verified by the test function.

References

  1. Borgonovo E., Plischke E., 2016. Sensitivity analysis: A review of recent advances. European Journal of
    Operational Research, 248, 869–887.
  2. Burnaev E., Panin I., Sudret B., 2017. Efficient design of experiments for sensitivity analysis based on
    polynomial chaos expansions. Annals of Mathematics and Artificial Intelligence, 81(1-2),
    187–207.
  3. Cheng K., Lu Z.Z., Zhou Y.C., 2017. Global sensitivity analysis using support vector regression. Applied
    Mathematical Modelling, 49, 587–598.
  4. Jansen M.J.W., 1999. Analysis of variance designs for model output. Computer Physics Communications,
    117, 35–43. Sobol’ I.M., 2007. Global sensitivity analysis indices for the investigation of nonlinear mathematical
    models. Matematicheskoe Modelirovanie, 19(11), 23–24.
  5. Saltelli A., Annoni P., Azzini I., et al., 2010. Variance based sensitivity analysis of model output. Design
    and estimator for the total sensitivity index.Computer Physics Communications, 181, 259–
    270.
  6. Janon A., Klein T., Lagnoux A., et al., 2014. Asymptotic normality and efficiency of two Sobol
    index estimators. ESAIM: Probability and Statistics, 18, 342–364.
  7. Hoef J.M.V., 2012. Who invented the Delta method?. American Statistician, 66(2): 124–127.
  8. Li W., Lu L.Y., Xie X.T., et al., 2017. A novel extension algorithm for optimized Latin hypercube sampling. Journal of Statistical Computation and Simulation, 87(13), 2549–2559.
  9. Jin R., Chen W., Sudjianto A., 2005. An efficient algorithm for constructing optimal design of
    computer experiments. Journal of Statistical Planning and Inference, 134(1), 268–287.
  10. Sudret B., 2008. Global sensitivity analysis using polynomial chaos expansion. Reliability
    Engineering and System Safety, 93(7), 964–979.