Optimization Of Transter Station Between Urban Rail And Bus Based On Smart Card Data

With the development of the national economy and the continuous improvement of urbanization level,more and more people gather to the metropolis,resulting in an increase in urban population and traffic demand.The contradiction between urban development and traffic is becoming more and more prominent.In this situation,prioritizing public transportation development becomes an efficient method to alleviate the increasingly serious traffic congestion problem.Rail transit and bus are two main modes of urban public transport system,which plays an important role in resident trip.So promoting the coordination between the two methods is the inevitable requirement of the development of urban public transport.The emergence of traffic big data provides a new way for solving urban traffic problems.Traditional data collection adopts the artificial methods,which has the characteristics of large workload,high cost,long duration and low data quality.With the increasing use of bus charging system and metro AFC system in public transportation system,the data obtained from the system is also used in every aspect of the urban public transportation.In view of the connection between urban rail transit and bus,based on the research on the present situation at home and abroad,this paper conducts the following researches on the basis of bus IC card data and metro AFC data:First of all,the characteristics of bus IC card data and metro AFC data are analyzed,and the coding information in the database is matched combining with the actual station information and line information.The data information redundancy and spatiotemporal information disorder of the metro AFC data and bus IC card are analyzed and fully cleaned to ensure the reliability of the basic data.Secondly,deviation data resulted from advanced-tagging behavior of passengers holding bus IC card are recognition and rectification.Advanced-tagging behavior is divided into two categories,unintentional advanced-tagging behavior and deliberate advanced-tagging behavior.Unintentional advanced-tagging behavior leads to the deviation of deal-time,so the up-time and on-station in the bus data are analyzed in order to get the actual arrival timetable,then the outliers generated by unintentional advanced-tagging behavior are recognized and rectification based on DBSCAN(Density-Based Spatial Clustering of Applications with Noise)algorithm and bus actual arrival timetable.Deliberate advanced-tagging behavior leads to the deviation of deal-time and off-station,and outliers generated by deliberated advanced-tagging are recognized based on transfer behavior and round trip characteristics,then the outliers are rectified.The method proposed in this study can be used to revise the bus data,so that the bus data can reflect the reality spatiotemporal information and the application value of the bus data can be enhanced,which can provides relatively accurate spatiotemporal data for the next passenger transfer behavior recognition and transfer station optimization model.Thirdly,the multiple stage journey of passenger on public traffic network is structured based on the recognition of transfer behavior.First,the concept of transfer is defined,depending on whether the activity between two continuous stages is the main purpose of the journey.Then,bus and metro stages were analyzed before transfer analysis.Third,the range of elapsed time thresholds of the three transfer modes including bus transfer metro(B-M),metro transfer bus(M-B)and bus transfer bus(B-B)are given by analysis.Finally,according to the given elapsed time thresholds,multiple stages in one journey is recognized and a complete journey of passenger on public traffic network is structuredFinally,bus stations in interchange network consisted of urban rail and bus are optimized based on above research.Localized network is described and topology structure is abstracted.The optimization model of transfer stations comprised of urban rail and bus is constructed,with the minimum transfer distance as the goal,the transfer volume and transfer node as the constraint,then the design procedure is used to solve the model.The interchange network composed of XIZHIMEN metro station and its surrounding bus routes is selected as a case study,and local transfer node in the interchange network is optimized.The result shows that transfer efficiency is improved and practicability of the model is verified.