Research On Emergency Derivative Network Based On Layered Cognitive Model

In recent years the emergencies are occurring frequently. A series of basic events derived from the initial incident to human society has brought great loss of life and property. Emergencies play an important role in the field of emergency management; the research about the derivative of emergencies has attracted the attention of academia. The limitations of the current study are mainly two points; first, most of the studies only study the relationship between emergencies from the numerical model or concept on the single-level research. Secondary, incidents may be caused by multiple events before, and existing researches pay more attention on the causal relationship between the two events.For the two limitations which are mentioned above, this paper study the level of the emergency with the knowledge unit theory, analyzing the formation mechanism of emergency derivative network and giving the general steps building the derivative network of emergency: first, the emergencies elements are discussed on the hazard bearing body layer, basic events layer and events described layer, building the knowledge unit model of hazard bearing body and the basic event. Then, the formation mechanism on the bearing body layer and basic event layer are analyzed, derivative relations are described and given the mapping method from the knowledge unit layer to the basic event layer. At the same time this paper proposes a method of building the emergency derivative network. At last, it takes storm events for example to show the application of this method.The significance of this study is obvious:it makes the decision-makers recognize the influence which is arising from the derivative network to analyze the occurrence and evolution of the emergency from the knowledge unit level, the rules level and the conceptual level, it also help those people make respond decision more quickly. At the same time the emergency layered model is more suitable for computer system design, providing a theoretical basis for the derivative network simulation on the computer platform.