In order to solve the problem of attribute redundancy in decision making,reduce the complexity of decision inference process, and realize the intellectualized railway emergency decision making with incomplete information,a new railway emergency decision making method has been proposed on the basis of rough set theory and Bayesian network.The minimal decision information set has been extracted by rough set knowledge reduction method with information entropy to realize the reduction of emergency situation information set,thus reducing the number of situation network nodes as well as the complexity of Bayesian network.The probability decision reasoning of railway emergency situation prediction has been achieved by the reduced Bayesian network model.Case analysis shows that the method not only meets the re-quirements of railway emergency decision but also is valid with incomplete information.%为了解决决策属性的冗余问题，降低决策推理过程的复杂性，实现在信息不完备情况下铁路应急决策的智能化，基于粗糙集理论与贝叶斯网络提出一种新的铁路应急决策方法。利用基于信息熵的粗糙集知识约简方法提取最小决策信息集，实现对应急态势信息集的约简，从而减少态势网络节点数目，降低贝叶斯网络的复杂性。基于约简后的贝叶斯网络模型实现了铁路应急态势预测的概率决策推理。案例分析表明该方法能够满足铁路应急决策需求以及在信息不完备条件下的有效性。
展开▼
