An Intercity Railway Dedicated Line Passenger Flow Analysis And Application System Development

With the operating of intercity passenger transport line,it had brought great convenience to people’s travel.However,due to the changes of demand for passenger travel,the passenger volume on the transport line was showed different distribution characteristics in time and space.The traditional methods for passenger volume analysis,such as qualitative analysis and graphic statistics,were generally used on extraction of single object,however,for a large amount of data with multiple objects,data mining was primarily used to extract the passenger volume distribution features.As an important method of data mining,clustering analysis can be applied to extract space-time distribution characteristics of passenger volume.It had significant guidance and practical value for solving these problems which include: optimizing the operating plan and capacity allocation,improving the science of passenger organization.In this paper,based on analyzing the development and the operating features of inter-city railway passenger railway in China with current general analysis methods for passenger volume,some analysis methods of clustering were studied in depth,in addition,it took some inter-city rail passenger line’s data as the research object,the space-time distribution characteristics were extracted,then,the information system which can automatically analyze and display the space-time distribution characteristics of this line was designed and developed.The main contents about the paper were as follows:(1)Research about spatial distribution features.Using hierarchical clustering analysis and K-means clustering,the spatial distribution features were respectively studied from three aspects,including the stations,the passenger volume between stations,the passenger volume of transportation line.The grades of passenger stations were re-marked according to the real passenger volume,it was found that it was more reasonable to set station G and station H as grade 2 and relegate station C from grade 2 to grade 3,the effect and reasonability of capacity allocation were also analyzed.At the same time,by analyzing the peak hours and passenger origin a day,it was found that the peak hours of station H were during 8:00-9:00am and 15:00-16:00pm,and the primarily passenger volume direction was station G.Taking the passenger volume between stations as research objective to analyze the correlation intension between stations,it was found that station A had strong association with station B,station G has strong correlation with station H;Letting passenger volume and transport density be indexes,it took some digging to get that the railway line was station H-centric,and most passengers in this line were short-distance passengers,what’s more,the passenger volume over station H was drop off one by one.Moreover,by clustering transport density,it was shown that the busy traffic periods were September and October.(2)Research about time distribution features.Using hierarchical clustering analysis and K-means clustering,the daily passenger volume features of this line were researched from different time span,such as day,week,month,then,the daily passenger volume in the line was presented like single-peak type,the daily peak period was between 7:00-8:00am,and the weekly peak period was between Friday and Sunday,the monthly peak period was between September and October.Based on the analysis of the passenger volume during the holidays,the distribution characteristics between stations and peak periods during weekends,3 days’ long vacation,the 7 days’ long vacation and the Spring Festival were extracted.(3)Developing about application system of passenger volume analysis.It described the overall framework of the system designing and the function modules of the system.At the same time,the designing and modeling of database were also made.What’ more,the core codes of hierarchical clustering and the algorithm K-means applied to analyze space-time distribution characteristics of passenger volume were sealed and developed,and the interface to display the result of analysis were also developed.