CBR-based dynamic decision-making method for emergencies
Cui Wang, Yinghua Song, Danhui Fang · IOP Conference Series Earth and Environmental Science · 2020
Abstract The randomness of emergencies is very important to emergency management, and case-based reasoning has attracted the attention of emergency decision-makers. In order to improve the one-sidedness of the existing emergency decision-making method based on static case-based reasoning, this paper fully considers the development and evolution stage of the event, and proposes a CBR-based dynamic emergency decision-making method. First, the emergency event is decomposed into multiple time stages; then the comprehensive attributes of each stage are determined according to the event description and treatment measures to calculate the attribute similarity; then the case similarity is calculated through the attribute similarity, and the decision-making plan of current event stage is formed by combining the treatment measures of the case with the highest similarity; at the same time, the current decision-making measures are involved in the case reasoning of the next stage. Finally, a high-rise building fire case in the city shows the feasibility of this method.