Urban Bus Passengers Origin And Destination Parallel Estimating And Spatio-Temporal Analysis Of Passenger Flow

Urban bus is one of the major modes of public transportation for residents’daily travel,which has many advantages such as environmental protection and high efficiency and is beneficial to solving traffic problems in large and medium-sized cities.During the process of giving priority to the development of public transportation,grasping the characteristics of spatial and temporal changes accurately in passenger flow information of vehicles,lines and stations can not only provide corresponding suggestions of route planning for urban resident trips,but also provide services for bus planning and dispatching.Furthermore,it provides a scientific theoretical basis for the rational allocation of urban public resources and the optimization and adjustment of urban spatial structure.In this paper,taking urban bus of Xiamen city as an example,the popular distributed storage and parallel computing technologies are used to calculate passengers’origin and destinations based on the data source from IC card data and vehicle GPS data.On the basis of obtaining passenger OD,analyzing spatial and temporal changes of passenger:flow of vehicles,lines and stations respectively.The main research contents and conclusions are summarized as follows:(1)Research on the key technologies of public transit big data.Based on the characteristics of big data to understand the specific performance of big data in the field of public transportation,this paper introduces the related theory and technology of distributed storage and parallel computing in detail,and provides important theoretical and technical support for subsequent research.(2)Parallel calculation of mass transit passenger OD.This paper proposes an algorithm for parallel estimation of bus passenger OD(Origin-Destination).This method uses distributed storage(HDFS,HBase)to implement partitioned storage of IC card data and GPS data.Furthermore,combined with the continuous travel chain and the historical travel rules of residents,the OD information of passengers is estimated accurately and rapidly based on MapReduce.The method verification and comparative analysis are performed based on the bus data in Xiamen.The comparison with the existing OD estimation methods indicates that the method proposed in this paper have a higher extraction rate and computational efficiency of passenger OD information.(3)Temporal and spatial analysis of city bus congestion degree.Combing the bus passengers OD and parallel computing technology(MapReduce)to calculate the full load ratio of bus vehicles and lines.Bus congestion degree of Xiamen city are studied according to the bus congestion level division on vehicle congestion degree.Based on the daily travel patterns of residents,the temporal and spatial analysis of congestion degree of bus vehicles and lines is conducted from three periods(i.e.high peak period,flat peak period and low peak period).The study found that the bus lines that connect Xiamen inside and outside of the island not only have similar congestion area during morning and evening peak but also have a higher number and congestion degree than the bus lines just inside of the island.(4)Temporal and spatial change analysis of bus passenger flow at bus stations.Based on parallel technology,this paper calculates the time-sharing of passenger flow at bus stations and discusses the time distribution characteristics of passenger flow in different date(i.e.working days,weekends and holidays)and different groups(i.e.elderly,students and adults).The nuclear density analysis of passenger flow at bus stations explores the overall spatial distribution of passenger flow at bus stations and the change of each time period.Furthermore,the passenger flow of hot spots is visualized and the underlying causes are analyzed.The study found that the bus traffic mainly concentrated in Xiamen Island.The passenger flow on early morning peak within the island showed an "λ" shape distribution,which evolved into a "--",shape distribution of the northeast-southwestward direction over time.SM City Square,Zhongshan Road and the railway station shopping district are always the hotspots for bus passengers.