Analysis And Application Of Vehicle Monitoring System Data Based On Hadoop Distributed Computing Platform

With the development of technology and economy,the automotive industry develops rapidly.As the car ownership continues to increase,many problems have revealed such as traffic congestion and frequent accidents,bringing many negative effects to people's live.In this situation,Intelligent Transportation Systems(ITS)which is based on the Internet technology comes out,devoting to solve the traffic problems caused by the vehicle increase.As the core of ITS,Vehicle Monitoring System records many vehicle indexes dynamically,accumulating generous vehicle monitoring data.Most vehicle monitoring data is in idle at present,this paper plans to make a data analysis to the historical data in order to explore the potential value.According to the characteristic of the vehicle monitoring data and the research of the related technology in data analysis,this paper uses Hive and Pig which are the subprojects of Hadoop to complete the analysis and application of the Vehicle Monitoring System data based on the Hadoop distributed computing platform,taking full advantage of the parallel processing performance.First of all,this paper completes the Vehicle Monitoring System to record vehicle indexes dynamically and accumulate generous vehicle monitoring data for the research subsequent.Then it designs a data analysis method based on Hive,including the design of theme,logic model,multi-dimensional analysis,granularity selection,metadata,and so on.It also completes another data analysis method based on Pig to achieve more flexibility,less development cost.As a supplement to the first one,this method has a less complexity and delay,contributing to consummate the analysis and application of the Vehicle Monitoring System data based on the Hadoop distributed computing platform.At last,this paper verifies the advantage of the scheme in generous vehicle monitor data analysis by setting up test bench,proving the effectiveness in traffic management and decision support for enterprise or government.