摘要In emergency decision making (EDM),it is necessary to generate an effective alternative quickly.Case-based reasoning (CBR) has been applied to EDM;however,choosing the most suitable case from a set of similar cases after case retrieval remains challenging.This study proposes a dynamic method based on case retrieval and group decision making (GDM),called dynamic case-based reasoning group decision making (CBRGDM),for emergency alternative generation.In the proposed method,first,similar historical cases are identified through case similarity measurement.Then,evaluation information provided by group decision makers for similar cases is aggregated based on regret theory,and comprehensive perceived utilities for the similar cases are obtained.Finally,the most suitable historical case is obtained from the case similarities and the comprehensive perceived utilities for similar historical cases.The method is then applied to an example of a gas explosion in a coal company in China.The results show that the proposed method is feasible and effective in EDM.The advantages of the proposed method are verified based on comparisons with existing methods.In particular,dynamic CBRGDM can adjust the emergency alternative according to changing emergencies.The results of application of dynamic CBRGDM to a gas explosion and comparison with existing methods verify its feasibility and practicability.