Study On Unconventional Emergency Scenario Reasoning Method The Cases-based

Unconventional emergencies are always unpredictable, complicated and destructive, which challenge the decision-making process. Making full advantages of collected information from well-structured case pools will certainly contribute to the rationality and scientific nature of the decision-making in unconventional emergencies, have the characteristics of inadequate precursor, complicated disposal and serious damage, which bring great pressure to the decision makers. Reasonable structured case,constructing structured case base, making full use of the information of the case to make decision, will help increase the rationality and the scientific nature of the unconventional emergency decision.Based on the nature of unconventional emergencies and shared information demands by emergency decision-making, this research explores the structured-cases-based decision-making in unconventional emergencies, employing multiple methods, namely scenario analysis, case-based reasoning technology and Bayesian networking analysis with a special focus on the guidance and suggestive role of existing emergency events. In order to make a large number of existing emergency event cases play the guidance and suggestive role, the paper researches the unconventional emergency scenario reasoning method based on the structured cases. The main contents are as follows:Firstly, it in-depth analyzes the nature of unconventional emergencies, and concludes as unpredictability, complexity, destruction and uncontrollability.Correspondingly, the decision-making process is also under high pressure, involves cooperation from multiple stakeholders with multiple targets and insufficient information. Based on the above observation, this research proposes the structure of information-demand in decision-making of unconventional emergencies.Secondly, to satisfy the information demands, this research proposes "case-event-scenario" hierarchical structure, and further reveals the scenarios-based knowledge. The scenario is composed of "dimension-attribute-characteristic", which is interpreted by different knowledge system at each level.Thirdly, this research also studies the relationship within scenarios, the driving forces and the relationship between them based on the system dynamics.Meanwhile, it further analyzes the scenario Bayesian network inference model using MATLAB technology in the case of Qingdao 22th November accident.Fourthly, this research develops methods matching the similarity of scenario-based unconventional emergency, including structural scenario links, semantics similarity, emergency tasks and structural actions. Qingdao 22th November and Dalian 30th June accidents are studied to verify the effectiveness.Finally, the emergency case management system is further developed, with the functions of information retrieval, decision support and intelligent learning. In addition, it points out the framework, structure and process of emergency decision model based on structured case system platform."Study on unconventional emergency scenario reasoning method the cases-based" may make beneficial exploration for "scenario-response" theory of emergency management and emergency response technology, and improves the decision-making level.