Study On The Optimization Model Of Shuttle Between Urban Rail Transit And Feeder Bus

Now, with the scale of city continuously develops, the urban population graduallyincreases in the same time, the urban transportation problems have become increasinglyprominent. To develop public transport has become the only way to solve block and to keepunimpeded, and to reduce traffic pollution of the city. Urban rail transit, as an important partof the urban public transportation, with its advantages of fast, punctuality, high-volume,and low-pollution, have been built or under construction by many domestic large andmedium-sized cities. Urban rail transit has broad development prospects in China. However,conventional public bus, as the main part of China’s urban public transport in the current, hasits advantages of the flexible-and-easy setting of line. With the intelligent transportationsystem in our country’s development, in the process of integration urban public transportationbuilt, how to combine the economic and technological advantages of the two modeseffectively, to foster strengths and circumvent weaknesses, and to achieve effective shuttle, isan important initiative to improve the overall efficiency of the city’s public transportationsystem.In this paper, under the conditions of network operation, the shuttle of special case ofconventional public bus-feeder bus and urban rail transit is the core research content,focusing on the departure interval of feeder bus. The modeling and optimization of theproblem is built with probability theory and mathematical statistics method, and the heuristicalgorithm is designed to solve the model. Before modeling, the theoretical method of shuttlebetween rail transit and feeder bus is introduced, the distribution of passengers’ walking timein the transfer is studied, to provide a theoretical basis and data support for modeling.In the first，theory of transfer between rail transit and feeder bus is analyzed, classifyingthe transfer, showing that if the two modes have small departure interval, it has no necessity tooptimize the departure interval. Then, from the point of view of the cumulative distributionfunction, with the sampling of walking time for passengers to transfer, Weibull function isused to describe transfer walking time distribution, with its parameter estimation andparametric tests.On the basis of this study, the operational coordination between the urban rail transit andfeeder bus in the flat peak period is determined to be core of the research. Based on the thinking of feeder bus unequal departure intervals, minimizing total transfer time ofpassengers in the study period, the model of the total transfer time is established and geneticalgorithm is designed to solve the model. In the last, with the research of the shuttle betweenXi’an Metro Line II and feeder bus line162, the model is verified.The results show that the Weibull distribution can describe the distribution of transfertime of passengers well; the optimization of feeder bus’ departure interval in the flat peakperiods is more meaningful; in the flat peak periods feeder bus’ranging with departureinterval is easier to implied and optimization effect is obvious. After the optimization oftransfer in the Huizhanzhongxin station, in the study period, the transfer time of a singlepassenger is reduced of104.7seconds, the total transfer time of passengers are decreased by27,365seconds, equivalent to7hours and36minutes. In this paper, the optimization ofshuttle between the two transport modes with the passengers total transfer time model andgenetic algorithm is effective and has some practicality.