Publication :


Construction of bootstrap confidence intervals on sensitivity indices computed by polynomial chaos expansion

Authors: S. Dubreuil, M. Berveiller, F. Petitjean and M. Salaün
Year: 2014
Journal: Reliability Engineering and System Safety.
Volume: 121, Pages: 263-275

DOI: 10.1016/j.ress.2013.09.011
OATAO: 9717

Sensitivity analysis aims at quantifying influence of input parameters dispersion on the output dispersion of a numerical model. When the model evaluation is time consuming, the computation of Sobol' indices based on Monte Carlo method is not applicable and a surrogate model has to be used. Among all approximation methods, polynomial chaos expansion is one of the most efficient to calculate variance-based sensitivity indices. Indeed, their computation is analytically derived from the expansion coefficients but without error estimators of the meta-model approximation. In order to evaluate the reliability of these indices, we propose to build confidence intervals by bootstrap re-sampling on the experimental design used to estimate the polynomial chaos approximation. Since the evaluation of the sensitivity indices is obtained with confidence intervals, it is possible to find a design of experiments allowing the computation of sensitivity indices with a given accuracy.

Bibtex citation :
Author={Dubreuil, S. and Berveiller, M. and Petitjean, F. and Sala\"un, M.},
Title={Construction of bootstrap confidence intervals on sensitivity indices computed by polynomial chaos expansion},
Journal={Reliability Engineering and System Safety},

INSTITUT CLEMENT ADER, 3, rue Caroline Aigle F-31400 Toulouse, France.
Livraisons : +33 (0)5 61 17 10 07
Secrétariat ICA: +33 (0)5 61 17 10 85 ou +33 (0)5 61 17 10 82
Fax : +33 (0)5 61 17 10 80