Developments and applications of Shapley effects to reliability-oriented sensitivity analysis with correlated inputs - Archive ouverte HAL
Pré-Publication, Document De Travail Année : 2021

Developments and applications of Shapley effects to reliability-oriented sensitivity analysis with correlated inputs

Résumé

Reliability-oriented sensitivity analysis methods have been developed for understanding the influence of model inputs relatively to events characterizing the failure of a system (e.g., a threshold exceedance of the model output). In this field, the target sensitivity analysis focuses primarily on capturing the influence of the inputs on the occurrence of such a critical event. This paper proposes new target sensitivity indices, based on the Shapley values and called "target Shapley effects", allowing for interpretable influence measures of each input in the case of dependence between the inputs. Two algorithms (a Monte Carlo sampling one, and a given-data algorithm) are proposed for the estimation of these target Shapley effects based on the $\ell^2$ norm. Additionally, the behavior of these target Shapley effects are theoretically and empirically studied through various toy-cases. Finally, applications on two realistic use-cases (a river flood model and a COVID-19 epidemiological model) are discussed.
Fichier principal
Vignette du fichier
paper.pdf (708.07 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03106452 , version 1 (20-01-2021)
hal-03106452 , version 2 (11-05-2021)
hal-03106452 , version 3 (18-05-2021)

Identifiants

Citer

Marouane Il Idrissi, Vincent Chabridon, Bertrand Iooss. Developments and applications of Shapley effects to reliability-oriented sensitivity analysis with correlated inputs. 2021. ⟨hal-03106452v1⟩
798 Consultations
440 Téléchargements

Altmetric

Partager

More