Traffic Flow Analysis And Service Evaluation Based On Multi-source Big Data Of Public Transportation

With the maturity of various sensor technologies and the continuous development of urban public transportation,many cities generate a large amount of public transportation data every day,such as smart card data,bus GPS trajectory,bus vehicle dispatch records,and taxi GPS trajectories,etc.It not only records the operation of urban public transportation but also records the spatiotemporal trajectory of passengers.Because the way of storing data in the city is more traditional,the existing public transportation data has problems such as the lack of passenger trip trajectory and the confusion of spatiotemporal information.It is necessary to integrate multi-source public transportation data,combined with in-depth analysis and mining of big data technology,in order to provide a reliable source of data for the study of urban public transport passenger flow and public transport service evaluation.This paper firstly uses Shenzhen smart card data,bus GPS trajectory and bus station line network as the foundation of data,using spatiotemporal data mining and spatial analysis technology,the method of matching the boarding station with the bus trajectory is constructed.According to the passenger trip characteristics,the passenger alighting station matching method is proposed.Then,based on the smart card ID,a public transportation trip chain model relied on individual trip records is established to restore passengers' trip chains,statistical lines and number of passengers traveling at the station.With the calculation of the passenger trip chain,combined with the map of Shenzhen city,using regular grid to extract significant areas of public transportation in Shenzhen.Based on the significant area of public transportation,the public transportation hotspot corridor is extracted.Finally,according to the temporal and spatial changes of the hotspot corridor,the analysis and research are carried out.This paper has established a complete technical process from data preprocessing,trip chain estimation,to spatiotemporal pattern extraction and analysis of public transportation demand.Finally,this paper analyzes the trip modes of public transportation in Shenzhen from two different dimensions of time and space,and explains the macro-temporal law of Shenzhen public transportation,and analyzes the difference between supply and demand of public hotspot corridors,it evaluates the public transportation service level of hotspot lines,and it also analyzes the travel demand generated by hotspot lines.By analyzing the trip demand generated by hotspot lines,combined with the lack of actual operation of existing bus lines and roads,using the spatiotemporal optimization decision-making method,the lines and station optimization improvement plan for public transportation is proposed.The research results of this paper have been applied to the related research completed by the Shenzhen Comprehensive Traffic Operation Command Center in 2017.,assisted traffic professionals to effectively analyze and understand the public transportation mode of Shenzhen,and has important theoretical and practical value for urban transportation planning research,and provides suggestions for improving the level of public transportation service in Shenzhen.