Abstract: When multiple emergencies occur simultaneously in the railway regional network, the emergency resource reserve may be insufficient. Combining the improved analytic hierarchy process with grey clustering analysis method, the priority of rescue in multiple accident points was proposed. A multi-objective programming model for railway emergency resource dispatch was established, which had the shortest total dispatch time, the least total dispatch costs and the minimum total penalty costs (the further loss caused by the failure to meet a certain type of emergency resources). Furthermore, the priority parameters were introduced into the proposed model to obtain the emergency scheduling scheme of multiple accident points, and the problem of competition-conflict of emergency resources among railway multiple accident points at the initial rescue stage was solved. Taking the road network within the jurisdiction of a railway bureau as an example for empirical analysis, the research results show that the optimal dispatching scheme from each emergency rescue base to each accident point can be obtained by solving the multi-objective programming mathematical model with genetic algorithm, which verifies the effectiveness and scientificity of the proposed method.

Key words: traffic and transportation engineering; railway emergency management; resource scheduling optimization; multiple accident points; priority; multi-objective optimization

摘要： 当铁路区域路网同时发生多起突发事件时，应急资源储备量可能存在不充足的情形。通过改进层次分析法与灰色聚类分析法相结合，提出多事故点的救援优先权问题。构建总调度时间最短、总调度成本最少和总惩罚成本（某种类的应急资源未能得到满足时而引起的进一步损失）最小的铁路应急资源调度多目标规划模型，把救援优先权参数引入到模型中，获得多事故点的应急调度方案，解决了救援初期多事故点间应急资源的竞争与冲突问题。以某铁路局辖区内路网为算例进行实证分析，采用遗传算法对构建的多目标规划数学模型进行求解，可获得各应急救援基地到各事故点之间的最优调度方案，该方法具有有效性和科学性。

关键词: 交通运输工程；铁路应急管理；资源调度优化；多事故点；优先权；多目标优化