Line sampling and Fuzzy Interval Analysis for the propagation of aleatory and epistemic uncertainties in risk models

Elisa Ferrario (1) , Nicola Pedroni (1) , Enrico Zio (1)

1 SSEC - Chaire Sciences des Systèmes et Défis Energétiques EDF/ECP/Supélec

In this paper, an advanced Monte Carlo (MC) simulation method, namely Line Sampling (LS), is considered in combination with Fuzzy Interval Analysis (FIA) for improving the sampling efficiency in the hierarchical propagation of hybrid (probabilistic and possibilistic) uncertainties through a model for the risk-based design of a flood protection dike. The proposed approach is compared to that of standard MC sampling showing superior accuracy and higher computational efficiency.

Domaines: Sciences de l'ingénieur [physics]

Elisa Ferrario, Nicola Pedroni, Enrico Zio. Line sampling and Fuzzy Interval Analysis for the propagation of aleatory and epistemic uncertainties in risk models. ESREL 2013, Sep 2013, Amsterdam, Netherlands. pp.1-8. ⟨hal-00838667⟩