考虑车流时空分布不均衡的货物列车开行方案优化研究

Freight Transportation Service Network Design Problem with Unbalanced Spatio-temporal Distribution of Wagon-flow

为适应货主动态需求和运到期限要求,采用动态车流组织方法进行编组方案调整、列车运行方案与车流挂线的综合优化.考虑车流时空分布不均衡性,结合基本运行图架构,运用时空网络描述车流组织动态过程,考虑车流组织过程时空连续性、车站与线路时空能力及运到期限等约束,在静态车流组织优化模型目标基础上考虑车流到后待解与编后待发的延迟费用,构建货物列车开行方案整数规划模型.基于K短路构造初始解,依据列车能力利用率阀值评估方案,从而调整车流改编方案搜索邻域解,设计模拟退火求解算法.实例研究表明,相比静态模型,所建模型的决策变量能更清晰地反映车流对车站线路时空能力的占用状况,优化方案能有效匹配车流时空分布不均衡性,有利于车流接续与流线结合.

To provide differentiate services to heterogeneous shippers, freight transportation service network design problem is introduced to integrate car classification, train make-up and train scheduling by dynamic wagonflow organization methods. Based on basic train diagram and space-time network, an integer programming model is proposed by further considering classification and connection delay costs in the static optimization model objective. Constraints include spatio- temporal continuity between adjacent organization phases for wagon- flow, station and line time- space capacity and delivery deadline. A simulated annealing algorithm is used to solve the model, constructing an initial feasible solution by K- shortest path, designing neighborhood search strategy to adjust wagon-flow reorganization according to train capacity utilization threshold. One practical example shows that, compared to the static model, wagon- flow unbalanced spatio- temporal distribution is addressed effectively with more clearly description to wagon-flow time-space capacity occupation by decision variables, which help to coordinate wagon-flow transfer between different trains.