Assessing and Managing Risks due to Natural Hazards

Authors

Michael H. Faber

Abstract

This paper presents recent developments on methodologies for the assessment and
management of risks due to natural hazards. First a review is made on system characteristics
commonly applied in the field of natural hazards risk management at large scale. Thereafter, a
general model framework is proposed for the representation of consequences and knowledge
based on recent developments within the Joint Committee on Structural Safety (JCSS). The
proposed framework explicitly accounts for two significant characteristics of systems, namely
indicators and conditional dependencies. This in turn suggests that when using this framework a
Bayesian statistical framework for risk assessment is advantageous and it is shown how generic
Bayesian risk models may be formulated to facilitate the analysis using Bayesian Probabilistic
Nets (BPN's). It is outlined and discussed how the proposed model framework may be applied for
risk management problems considering natural hazards at large geographical scales. Finally an
example is provided on the application of the framework based on recent and ongoing research on
management of risks due to earthquakes.

Published in/by

Proceedings ISGSR2007, First International Symposium on Geotechnical Safety & Risk, Shanghai, China, October 18-19, 2007.