The project is supported by the U

Bayesian updating in natural hazard risk assessment

Additional experiments, to validate this approach and to develop it further, are being conducted by our collaborating organizations in South Korea. In general, two types of deterioration mechanisms can be identified, shock deterioration hazard-related, e. Then the areas that could be affected by the hazard are marked, this is called Hazard Mapping.

Hazard assessment helps us to identify the threats and understand their nature and behavior so that we can plan and prepare for the upcoming disasters. To realize real-time, automated condition assessment, the Bayesian updating efficiency needs to be improved. Social and political trends - Changes in policies, Re-locations of people, Conflicts. For many mathematical models, writing down a subjective prior can be a lot easier than trying to determine an objective prior. Natural factors - changes in the pattern of weather leading to new hazards like drought, frequent and extreme flood events.

However, before this detailed study it is necessary to establish how susceptible population groups are to the event and how vulnerable they are to this hazard. Hazards could include earthquakes, volcanic eruptions, floods, drought, cyclones and epidemics. Industrial hazards - chemical accidents, poisoning. These are the first steps in vulnerability analysis, and they are needed before any detailed hazard analysis. In the end, real-time risk assessment tools will be developed to enable emergency response manager for risk mitigation planning and decision making.

To realize realtime automated condition assessment

The objective of the present paper is the demonstration of the potential and advantages of Bayesian networks for the application in risk assessments for natural hazards. The requirements to improve the performance during one natural hazard often conflict with the corresponding requirements for another natural hazard. Earthquakes, for example, require different instruments and specializations for analysis than e. Figuring out how to do that is leading us on an interesting journey thru the statistical literature, and round the world, and is even leading to some new mathematical results.

Hazard assessment helps us

Then identify the factors influencing the hazards, e. The flooding of the Fukushima-Daiichi plant, the subsequent release of radioactive material, and the potential for similar events along the U. Bayesian updating provides a rational framework for supporting these tasks, with one of the biggest challenges being the computational difficulty in sampling from posterior distributions. Depending upon the progress of this work, it can be quite useful in the aging management and life extension of nuclear plants. If there is concrete prior information, the prior can be based on that.

These two together are often enough to determine a prior. The potential complexity of infrastructure models increases this challenge and hinders automated updating of condition and guidance of maintenance actions. The next step is to estimate or calculate the scale strength, magnitude of the hazardous event, also on an ordinal scale. The current scope includes smoothed particle hydrodynamics method and software package. This research will investigate the integration of soft and high-performance computing and advanced simulation techniques for establishing versatile stochastic optimization algorithms.