Probabilistic Analysis of Large-scale Engineered Systems using GIS

Over the last 3-4 decades significant progress has been achieved in the general research area of risk and reliability analysis. The link between risk and reliability analysis results and decision making has been strengthened and it is increasingly recognized that probabilistic analysis in engineering shall be brought into the context of decision making. At the present time it would be fair to say that the decision theory is recognized to comprise the basis framework for probabilistic risk and reliability in the area of structural engineering. Whereas just one decade ago the main emphasis of research efforts in the area of risk and reliability analysis in structural engineering was oriented on aspects of establishing efficient methods for the analysis of probabilities of failure of components and systems of relevance for design and assessment of structures, the research direction has shifted somewhat over the later years to focus on the aspects of modeling. Moreover, over the last decade it has been realized that structural systems cannot be considered and/or modeled in isolation; the context in which the structures serve a function has to be included when risks and reliabilities of structures and structural components are assessed. As a result of this shift of research orientation one profound new problem has emerged in the area of structural risk and reliability analysis, namely the probabilistic analysis of large systems. When systems are very large and when the systems exhibit performances which significantly are influenced by causal and statistical dependencies of events relating to the constituents of the systems then the analysis of the probabilistic characteristics of the system performance, as required in the context of decision analysis becomes numerical involving. Furthermore, when considering realistic engineered systems much of the information required for probabilistic systems analysis is not available and the probabilistic assessments must be able to account for this lack of knowledge. The aim of the present project is to aim to push the presently ongoing research in this area further, with special consideration of two interrelated problems; life cycle risk management of individual but large structures like bridges and tunnels and risk management of geographically distributed engineered systems, such as electricity distribution systems.

• Probabilistic Analysis of Large-scale Engineered Systems using GIS (ETH Research Database 19262)