基于大数据的城市智能公交管理系统的设计与实现

发布时间：2018-05-31 08:30

本文选题：城市公交 + 智能公交　；参考：《长安大学》2017年硕士论文

【摘要】：大力发展城市公交,实现城市公交智能化,是当前公认的解决城市交通问题的有效办法。目前我国部分城市已经引入了电子站牌、电子站台等智能手段,但是整体智能化水平仍然不高。特别是运营积累的海量公交数据具有明显的大数据特征,但基本没有得到充分的发掘及应用。有鉴于此,本论文围绕传统的城市公交管理业务,设计并开发了一款面向大数据的城市智能公交管理系统。该系统具有公交智能调度、乘客出行服务与数据统计分析等功能,对于充分挖掘公交数据价值、提升公交管理智能化水平具有重要价值。具体而言,本论文的研究内容包括:(1)系统总体设计。通过对我国城市公交行业城市公交运营和管理需求进行分析,设计了基于大数据的城市智能公交管理系统的整体框架。该框架围绕公交大数据的采集、传输、存储分析以及展示等方面进行搭建,涵盖了公交智能调度、到站时间预测和客流量预测等功能。(2)基于大数据的数据分析算法设计。通过利用MapReduce分布式计算框架对算法进行改造,从而将复杂的数据分析处理过程解构为相互独立的细粒度过程,并由独立的大数据处理节点进行并行处理,从而满足公交大数据的快速数据分析需要。实验证明,利用MapReduce改造的数据分析算法可以大大缩短数据处理时间,从而很好地满足城市智能公交管理系统对公交大数据的处理要求。(3)系统研发。为高效地进行系统开发,借用Jeesite基础架构,设计了基于JeeSite的城市智能公交管理系统。该系统采用分层架构,自下而上分别包括数据库层、数据访问层、业务逻辑层和展示层。并围绕数据分析流程,分别实现了相应的的大数据采集方案、数据处理算法以及数据分析结果的查询与展示模块。系统测试和模拟应用表明,该系统有助于挖掘公交大数据价值并提升城市公交的运营和管理水平。
[Abstract]:It is an effective way to solve urban traffic problems to develop urban public transportation and realize urban transit intelligentization. At present, some cities in our country have introduced intelligent means such as electronic station board and electronic platform, but the overall intelligence level is still not high. Especially the mass transit data accumulated by operation has obvious big data characteristics, but it has not been fully explored and applied. In view of this, this paper designs and develops a big data oriented urban intelligent bus management system around the traditional urban transit management business. The system has the functions of bus intelligent dispatch, passenger travel service and statistical analysis of data. It is of great value to fully excavate the value of public transportation data and improve the intelligent level of bus management. Specifically, the research content of this paper includes the overall design of the system. Based on the analysis of the operation and management requirements of urban public transport in China's urban transit industry, the overall framework of urban intelligent bus management system based on big data is designed. The framework is built around the collection, transmission, storage analysis and display of the bus big data. It covers the functions of bus intelligent dispatch, arrival time prediction and passenger flow prediction, etc.) the data analysis algorithm based on big data is designed. By using the MapReduce distributed computing framework to transform the algorithm, the complex data analysis and processing process is deconstructed into an independent fine-grained process, and parallel processing is carried out by an independent big data processing node. In order to meet the bus big data rapid data analysis needs. Experimental results show that the data analysis algorithm modified by MapReduce can greatly shorten the time of data processing, and thus meet the requirements of urban intelligent bus management system for the processing of bus big data. In order to develop the system efficiently, the urban intelligent bus management system based on JeeSite is designed by using Jeesite infrastructure. The system adopts hierarchical architecture, including database layer, data access layer, business logic layer and presentation layer from bottom to top. Around the data analysis flow, the corresponding big data collection scheme, data processing algorithm and query and display module of data analysis results are implemented respectively. The system test and simulation show that the system is helpful to excavate the big data value of public transportation and improve the operation and management level of urban public transportation.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U495;TP311.13

【参考文献】