Analysis Of Vulnerability Based On Compound Extreme Value Theory And Fault Tree Analsis

With the development of society and economy, and environmental disruption, all kinds of emergency incident increase gradually, including the man-made and the natural. These emergency incidents will not only cause heavy casualties and huge economic losses, but also impact on social stability seriously. Domestic and foreign scholars generally agree that the loss of emergency incident is caused by two reasons: one is the magnitude of emergency incident's destructive power; the other is the vulnerability degree of the affected object. The purpose of this dissertation is to study the method of vulnerability analysis about emergency management. The dissertation can be divided into three parts and each of them as follows:First, this dissertation proposes a theoretical framework of vulnerability analysis, based on a summary of relevant research. This theoretical framework proposes an analysis of vulnerability mainly from the measurement of vulnerability and the recognition of vulnerability factors.Second, a compound extreme distribution model is developed in this dissertation to analyze hazard. With this model, we can forecast the probability of disaster, the return period and the losses, which are very important index of vulnerability. In this dissertation, using Bradford Disaster Scale as a starting point, based on religious mathematics theory, we develop a compound extreme distribution model, which is composed of emergency incident's frequency and the losses caused by it as variables. And this model can provide theoretical basis for the establishment of emergency preparedness system. What more, though the analysis of this model we can conclude that: the probability of disaster above some scale is inverse proportion to emergency incident's frequency.Third, a method of vulnerability analysis is presented which is based on Fault Tree. According to the factors which should be considered in the analysis of vulnerability and the characteristic of Fault Tree analysis (FTA), this dissertation proposes an analysis of vulnerability based on FTA. And we propose an algorithm of minimum cut of fault tree based on step function, which is easier to understand. And an example is given to demonstrate computation process. We can choose major vulnerability factors by the algorithm of minimum cut, and analysis the affect of these factors to vulnerability.Last, give an empirical analysis of vulnerability in meteorological disaster. We can get economic level, population density and environmental protection are the major factor that affect meteorological disaster vulnerability.