Implementation Of A Big Data-based Intelligent Urban Transit Management System

It is well recognized that developing intelligent urban transit system is an effective way to ease the traffic problems in cities.Nowadays,although some intelligent devices,such as electronic bus-stop board,intelligent bus station,and so on,can be spotted in some cities in China,the overall level of intelligence in urben transit is still far from maturity.In particular,the mass transit data accumulated in daily operations has already possessed the 4V characteristics of big data,but with values barely exploited and utilized.On this account,by focusing on the business aspect of traditional urban transit management,this paper designed and implemented a big data-oriented urban intelligent transit management system.The system has integrates functionalities like intelligent bus dispatching,passenger travel service and data statistical analysis,andcan play an important role in fully exploiting the value of public transportation data,as well as in enhancing the intelligent level of bus management.Specifically,major research contents of this thesis can be summarized as follows:(1)System framework design.After summarizingthe requirements of urban public transport operations and management,the overall framework of the bigdata-based urban intelligent transit management systemwas designed.This framework covers aspects like collection,transmission,storage analysis and display of big data,and containsmajor businesses like bus intelligent dispatching,arrival time forecasting,passenger volume forecasting,and so on.(2)Design of big data-based analysis algorithms.These algorithms are combinations of both MapReduce porcesses and the theories of data analysis algorithms,and then the complex data analysis process is divided into independent fine-grained process,which is handled in parallel by separate large data processing nodes.The experiments results show that these algorithms can greatly shorten the data processing time,therefore meet the requirements of big data processing.(3)System implementation,test and application.As an efficient way of software development,Jeesite,a light-weight Jave EE platform,is used in the system implementation.As a result,this system has a hierarchical architecture,including layers for database,data access,business logic,as well as presentation.In addition,other key parts of this system are like relevant big data collecting services,data analysis algorithms,as well as modules for query and display of data analysis results.Finally,throughout test and demonstrating application,it can be concluded that this system is capable of mining values from the huge volume of transit data,and therefore are of great value in promoting the operation and management of urban transit systems.