Derivation And Application Research Of Bus Passenger Flow OD Matrix In Big Data Environment

With the rapid development of urban public transport in China,a large number of intelligent technologies are widely used in public transport.The use of these technologies,not only to facilitate the daily operation of public transport enterprises,but also to collect a large amount of data resources for the bus company.Through the analysis and processing of these data resources,the bus companies can derive and analyze information such as bus passenger flow,and provide valuable decision support for bus operation,management and planning.However,in face of the cumulative and increasing bus data,the traditional data processing methods and algorithms gradually powerless.In order to explore and apply the bus data better,it is urgent to use the methods and ideas of big data to analyze and process the bus data.In view of the above problems,this paper designed a big data analysis platform based on Hadoop architecture,to meet the processing needs of mass transit data,combined with big data thought and data situation,put forward the OD matrix derivation algorithm and the application method of bus passenger flow data.Compared with the idealization and simplification of the previous research,this paper takes the IC data,bus GPS data and bus scheduling data as the basis,taking into account the problems of the actual bus data,in the passenger derivation process to correct and complete the data.Taking into account the past passenger data application method is too unilateral and thin,this paper proposed a way to comprehensively explore the application of bus traffic data.Firstly,this article introduces the relevant background knowledge,the basic theory and the technical profile used in this study were described.Secondly,the paper analyzes the data of the bus system and preprocesses the data,based on the analysis of the bus data and the demand of the bus service,the paper designs and constructs the bus data analysis platform,use HDFS,MapReduce,Hive and other technical analysis to deal with the bus data.Thirdly,based on the big data processing environment,proposes an algorithm to deduce the OD matrix of bus traffic,using IC data,bus GPS data and bus scheduling data.Respectively,from the judgement of get-on site,judgment of get-off site,transfer analysis and other steps to explain.Finally,combined with the needs of bus companies,the paper puts forward the theme mining and application methods of passenger flow data,analyzes and applies the data from the site traffic analysis,regional traffic analysis and the traffic matching analysis respectively,and illustrate the methods with instances.