Train Operation Organization And Optimazation In Urban Rail Transit Network

Some big cities in the developed countries, such as New York, Paris, Tokyo and Moscow, have already entered the era of Urban Rail Transit Network management and operation, and they also have rich relevant management experience. However, China’s urban rail transit operation is still in the development stage. Its relevant management experience is still not enough and some operation methods even lag behind those big cities. Urban rail transit lines in Shanghai, Beijing, Guangzhou and Nanjing have basically come into a network form. At the same time, the main urban rail transit lines in Hangzhou, Tianjin, Chengdu and other cities have been built too. It is important to establish different train marshalling plans and routing modes which can adapt themselves to both operation facilities and passenger flow distribution so as to ensure safety, improve operation efficiency, reduce travelling time and lower the cost.Based on the background above, firstly, the basic structure of the urban rail transit network and passenger flow characteristics were studied in the thesis. Then a Gray BP model was used to predict the largest passenger flow volume during the peak hours. Secondly, many factors influencing the train formation in urban rail transit were analyzed and different types of formation were discussed too. Based on the Analytic Network Process (ANP), a model was built to compare the advantages and disadvantages of three different train formation schemes. Thirdly, different types of turn-back tracks and stations were described in the thesis so as to set a base for the optimization of train routing in urban rail network. Finally, a model to minimize the light vehicle turnover was also built to find out the ideal train marshalling plan and routing modes so as to optimize the whole operation. A representative example of urban rail transit network was then designed to verify the practicality of the model. In the example, different marshalling plans and routing modes were analyzed and their corresponding objectives were found. A Matlab program was then designed to improve the practicality and efficiency of the model.