铁路非常规突发事件应急救援决策方法研究

Decision-making Method for Unconventional Railway Emergency Rescue

将贝叶斯网络推理方法应用到铁路突发事件应急决策中,为铁路突发事件应急救援提供一种实用、智能的科学辅助决策方法.在提取、约简案例决策属性的基础上,采用基于云模型的评估方法完成案例决策属性定性划分,降低决策属性划分的偏差性.构建铁路突发事件应急决策贝叶斯网络模型,并提出进行综合决策的具体方法及步骤.通过具体算例说明使用贝叶斯网络进行铁路非常规突发事件综合概率的决策过程.案例分析表明,该方法能够满足铁路应急决策需求,在信息不完备情况下同样具有实用性和有效性.为领域知识不完全,主要依靠经验知识进行铁路突发事件应急决策支持系统的智能化提供了新方法、新思路.

The inferential method of Bayesian network is applied to the railway emergency decision-making to provide a practical, intelligent and scientific decision-making method for the railway emergency rescue. On the basis of the extraction and reduction of decision attributes, the assessment method based on cloud model is used to qualitatively divide the decision attributes of the case, which reduces the deviation of the division. The Bayesian network model of the emergency decision-making for railway emergency is established, and a specific measure of comprehensive decision-making is proposed. The example shows the decision-making process of comprehensive probability of unconventional railway emergency. The analysis of the case shows the measure can meet the need of railway emergency decision-making, which is practical, precise and valid even when the information is incomplete. The proposed measure can provide new ideas to intelligentize railway emergency decision-making support system while the knowledge of field is incomplete and the experience is dominating.