Abstract: Railway terminal is generally located at the intersection of multiple railway main lines and branch lines, consisting of marshalling yards, loading and unloading station, linkage lines and some transportation service equipment, which is an important part of railway transportation network. It mainly undertakes the departure and arrival of wagon flow from railway network, pickup and delivery of local wagon flow, and efficiently realizes the goal of transship among train flow, wagon flow, and cargo flow. The wagon flows in the railway terminal mainly include through wagon flow, local wagon flow arriving with inbound train, transship wagon flow arriving with inbound train, and local wagon flow departing with outbound train. They share the technical equipment and resources in railway terminal. Therefore, the reasonable work division of marshalling yards can directly reduce the detour routing and repeated operation of wagon flow in the terminal. And the unreasonable pickup and delivery of local wagon flow will drop. The various technical operations arisen from railway terminal will appear more coordinated. The efficiency of the railway transportation network will be promoted.
Marshalling yards are the hub of railway terminal. In large railway terminal, there are often many railway lines connecting to different railway directions. And the pickup station delivery station of local wagon flow is scattered. The direction of outbound train coupling transship wagon flow is complicated. To meet distribution rule of wagon flow, avoiding generating angular wagon flow to increase the workload of railway terminal capacity, two or more marshalling yards often are arranged into the railway terminal. The vast majority of large railway terminal around the world is equipped with two or more marshalling yards. Although the number of marshalling yards in railway terminal is relatively small in China, there are still more than 20 railway terminals equipped with two or more marshalling yards.
The integrated optimization for train and wagons in railway terminal with multiple marshalling yards is studied. It is framed as a mathematical model that incorporates certain constraints which reflect the railway directions connecting marshalling station, handling stations connecting marshalling station, the capacity of train arrival-breakup-accumulation-makeup-departure, and the capacity of transshipping. It intends to minimize the overall cost of inbound train moving, wagon reorganization and accumulation, and local car shunting. According to the specific structure of the programming model, an integrated approach with two stage combining greedy generating procedure and asynchronous iteration heuristic is described as IA-TS. Firstly, a greedy procedure is proposed to arrange the destination marshalling station of the main wagon group to the train. Then the scheme matching the train and marshalling station is updated with the secondary wagon group which is ranked second in number of wagon groups. And then the matching scheme is verified with the capacity of train arrival, train breakup, wagons accumulation, train makeup, train departure, and wagons transship. So, the feasible matching scheme set can be generated. Secondly the matching scheme is coded using the natural integer sequence. An interactive-inherent updating procedure of filtered matching set is offered to discover the best matching scheme. The performance of the proposed approach is evaluated and compared to other algorithms via the testing of different-sized test cases.
In this study, the multiple marshalling yards is homogeneous in the railway terminal. They are not classified into main and auxiliary marshalling yards. The future research will focus on the railway terminal under heterogeneous operation of the main and auxiliary marshalling yards.

Key words: railway terminal, marshalling yard, handling station, train flow, wagon flow

摘要： 研究一类面向多编组站铁路枢纽的列流-车流协同组织优化问题。以列车进出站走行、货车改编与集结、地方作业车取送费用最小为目标,考虑枢纽各编组站接入方向、衔接装卸站情况、场站到解集编发能力和枢纽转站能力等限制构建模型,并根据模型特点设计两阶段融合求解策略。该策略首先给出主核心车组接入编组站作为列车接入站的贪婪匹配方法,进而利用次核心车组进行方案调整,然后根据到解集编发能力和转站能力限制完成匹配方案检验,得到可行匹配方案集。其次,给出列车与编组站相互匹配的编码方案,进而利用基于优选集的交互-内生更新过程找出最优列车-编组站匹配方案。最后通过不同规模算例测试所提方法的求解性能。

关键词: 铁路枢纽, 编组站, 装卸站, 列流, 车流