Bayesian networks and influence diagrams for earthquake risk management

The recent series of natural hazards (Sumatra earthquake 2004, Hurricane Katrina 2005) not only lead to an improved awareness, but also highlighted the difficulties involved in an efficient decision making process in such situations, especially in less developed countries. Consistent and quantitative risk assess­ment tools for buildings and infrastructure in seismic active areas are urgently needed to ensure an efficient decision making process that facilitates the optimal allocation of available eco­nomical resources for the management of risks.

Research project

This project constitutes the fundamental part of the interdisciplinary research project MERCI (Mana­ge­ment of Earthquake Risk using Condition Indicators) at the ETH Zurich aiming to develop an earth­quake risk management and decision support system, addressing the specific decision situations for the bodies responsible for a group of structures before, during and after an earthquake. A metho­dology of generic nature is established and is assumed to be applicable to different regional charac­teristics. In order to achieve this, the concept of condition indicators is investigated and further deve­loped.

The theoretical framework for the risk management is the Bayesian decision theory. Risks are quantified using influence diagrams or Bayesian networks utilizing indicators (see Figure). As a first activity the decision problems for the three different decision situations are identified and formulated such that they may be represented and assessed individually in prior decision analysis for the purpose of identifying activities for efficient risk reduction. Furthermore they are also assessed by means of pre-posterior decision analysis for the purpose of identifying how additional information may efficiently reduce the risks. A uniform basis for the representation of the uncertainties dominating the decision problems is developed and specified.

Bayesian network for the Value of Information analysis in the context of earthquake risk management.

Prof. Michael H. Faber, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Yahya Y. Bayraktarli, This e-mail address is being protected from spambots. You need JavaScript enabled to view it

ETH Zurich

Department of Civil, Environmental and Geomatic Engineering, Institute of Structural Engineering

Chair of Risk and Safety, http://www.ibk.ethz.ch/fa/index_EN

MERCI project homepage, www.merci.ethz.ch