大规模出租车OD数据可视分析系统

发布时间：2018-10-30 06:32

【摘要】：随着城市化的发展和城市人口的增加,现代大都市面临着严重的城市问题,如交通拥堵和不理想的城市资源规划等。近年来,随着技术的快速发展和不同的传感技术的广泛应用,大量的数据已成功收集并且记录了城市的各个方面的状态。这给解决都市问题和创建基于城市大数据分析的智能城市带来了新的机遇。本文主要针对出租车GPS数据进行研究,让交通运管部门预知各区域出租车需求的变化情况,对出租车进行更合理的规划。本文研究了出租车OD(Origin-Destination)数据的时空分布特性,对出租车起点终点特征数据分别进行分析,并且通过热度图和聚类图,引导用户进一步地分析感兴趣以及相对有特点的区域。结合环形像素图,时空堆栈图,柱状像素图等专门针对于出租车OD数据进行分析的可视化组件,让用户从交互探索中得出城市发展各个区域内出租车流向的差异性,以及从出租车数据中挖掘不同日期以及一些城市规划事件对人们出行模式的影响。本文的可视分析系统利用当下流行的Hadoop平台来对大规模的出租车GPS数据进行管理、预处理以及提取我们需要的出租车OD数据。系统采用B/S架构的方式将各个可视化组件作为前端数据分析手段,为了加快用户在使用可视化组件分析数据时的响应速度,本文采用网格索引对出租车OD数据进行查询加速,交通运管部门利用我们的可视分析系统可以及时发掘出租车聚集模式,提前管理和调度车辆。
[Abstract]:With the development of urbanization and the increase of urban population, modern metropolis is faced with serious urban problems, such as traffic congestion and unsatisfactory urban resource planning. In recent years, with the rapid development of technology and the wide application of different sensing technologies, a large number of data have been successfully collected and recorded the status of various aspects of the city. This brings new opportunities to solve urban problems and create intelligent cities based on urban big data analysis. In this paper, the GPS data of taxis are studied, so that the traffic management department can predict the changes of taxi demand in various regions, and make more reasonable planning for taxis. In this paper, the spatial and temporal distribution of taxi OD (Origin-Destination) data is studied. Guide users to further analysis of areas of interest and relative characteristics. Combined with annular pixel map, space-time stack map, columnar pixel map and other visual components which are specially used to analyze taxi OD data, users can find out the difference of taxi flow in different areas of urban development from interactive exploration. And mining different dates from taxi data and some urban planning events on the impact of people's travel patterns. The visual analysis system in this paper uses the current popular Hadoop platform to manage, preprocess and extract the taxi GPS data that we need. In order to speed up the response speed of users when using visual components to analyze data, the system adopts the method of B / S architecture to take each visual component as the front-end data analysis method. In this paper, we use grid index to speed up the query of taxi OD data, and the traffic management department can use our visual analysis system to discover the taxi gathering mode in time, and to manage and dispatch the vehicle ahead of time.
【学位授予单位】：浙江工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495;TP311.13

【相似文献】