Emergency decision-making model of environmental emergencies based on case-based reasoning method.

Environmental emergencies are characterized by high uncertainty, complex evolution, and potential for serious damage, thus posing enormous pressure and difficulties to the emergency responses of enterprises and governments. Improving the efficiency and quality of emergency decision-making constitutes the primary focus of today's research in this field. This study systematically analyzes the scenario evolution mechanism of environmental emergencies with a multi-dimensional scenario space method, and key scenario factors are identified from disaster-inducing factors, disaster-bearing factors, disaster-pregnant environments, and emergency actions. Based on these, an emergency decision-making model for environmental emergencies (EEEDM) is constructed based on case-based reasoning (CBR). First, different matching algorithms are designed for accurate numerical data, fuzzy semantic data, and symbolic data. The similarity between the target scenario and the historical scenario is calculated, and the historical scenario similarity set is built according to the given threshold value. Finally, the emergency action plan of the scenario is modified with its utility value evaluated. A solution that applies to the target scenario is then obtained. Additionally, the decision-making model proposed in this paper is validated by an example of environmental emergencies. The results show that this model is scientific and reasonable, and it can better realize the multi-dimensional expression and fast matching of the scenarios and meet the decision requirements of "scenario-response". In practice, the model is capable of providing support for relevant departments' emergency decision-making.