The Management of Na-Tech Risk Using Bayesian Network

Abstract

In the last decades, the frequency and severity of Natural-Technological events (i.e., industrial accidents triggered by natural phenomena or Na-Techs) increased. These could be more severe than simple technological accidents because the natural phenomenon could cause the prevention/mitigation/emergency systems fail. The dynamic assessment of the risk associated with these events is essential for a more effective prevention and mitigation of the consequences and emergency preparation. The main goal of this study is the development of a fast and dynamic tool for the risk manager. An approach supporting the management of the consequence is presented. It is based on the definition of a risk-related index, presented in the form of a discrete variable that combines frequency and magnitude of the events and other factors contributing to the worsening of Na-Tech. A properly designed Geographical Information System (GIS) allows the collection and processing of territorial information with the aim to create new data contributing to the quantification of the Na-Tech risk index. A Bayesian network has been built which efficiently lends in including within the model multiple elements with a direct or indirect impact on the distribution of risk levels. By means of this approach, a dynamic updating of the risk index is made. The proposed approach has been applied to an Italian case-study.

Keywords

• natural-technological events
• hazardous material
• chemical industry
• dynamic risk assessment
• risk management
• flood