基于约束程序累积调度的编组站静态配流模型研究

Study on Static Wagon-flow Allocation Model of a Marshalling Station Based on Constraint Programming Cumulative Scheduling

静态配流是铁路编组站阶段计划的核心,模型和算法的优劣直接影响编组站作业效率和经济效益。本文基于约束程序累积调度和字典序多目标优化理论,考虑配流成功的出发列车优先级总和最大、出发车流来源总数最小、到达车辆先到先发等具有字典序的3个目标,以满轴、正点、不违编、解编顺序及编组场容量限制等为约束条件,建立静态配流字典序多目标累积调度模型。采用迭代、约束传播和回溯算法求解。通过现场实际数据验证:本算法求解时间满足现场要求;模型稳定、扩展性好,符合实际需求。

Static wagon-flow allocation is the core of the stage plan of a marshalling station, its model and algo-rithms being good or bad affects the operating efficiency and economic benefit of the station directly. In this pa-per, in accordance with the theory of constraint programming cumulative scheduling and lexicographic multi-objective optimization, the static wagon-flow allocation lexicographic multi-objective cumulative scheduling model was set up for the purpose to maximize the sum of departure trains of preference, minimize the sum of sources of departure trains and keep the rule of the first to arrive and the first to depart for arrival trains under constraint that departure trains must be punctual, with full axis, not against train grouping requirements, and following sorting and formation order and limitation of capacity of the shunting yard. The model was solved with the iteration method, constraint propagation and backtracking algorithm. Site data verify that the model is stable realistic and of good augmentability and the solution time of the algorithms meets the site requirements.