基于出租车GPS数据的高效益寻客推荐策略研究

发布时间：2018-03-23 17:48

本文选题：出租车GPS数据　切入点：时空分析　出处：《浙江大学》2017年硕士论文

【摘要】：随着经济的快速发展和城市化建设的日益推进,人们对出行服务的需求愈发增长,出租车作为城市公共交通的重要组成部分,在人们的日常出行中发挥着重要作用。然而,因客源时空分布不均衡等因素而导致出租车高空载率、争抢客源等不良现象频现,大大增加了出租车寻客的难度,降低了出租车的运营效率。随着ITS(Intelligent Transport System)技术的发展,越来越多的学者将出租车寻客模式作为研究热点,但普遍缺乏出租车效益评价指标和顾及出租车供需关系及交通状态的推荐策略方法研究。本文以深圳市13798辆出租车GPS数据为研究对象,以提高出租车效益为研究目标,在分析影响出租车寻客效益指标因素的基础上,充分考虑推荐的高效益载客热点区域出租车的供需关系及交通状态,提出了基于出租车GPS数据的高效益寻客推荐策略方法,为出租车高效益寻客提供辅助决策支持,同时对于改善出租车资源分布不均衡的现状,充分发挥出租车在城市交通中的补充作用有着重大意义。通过分析影响出租车寻客效益的指标因素,本文以出租车载客态及相邻空载态相结合作为研究对象,提出了基于出租车载客态单位时间收入及相邻空载态寻客时间的效益指标模型。基于该模型,在对出租车GPS数据进行数据预处理及地图匹配的基础上,采用经典数理统计及地统计分析方法,对出租车高效益客源的时空分布特征进行了研究。通过对比分析工作日与非工作日及同一天不同时段的客源量变化特征来间接反映高效益客源随时间在数量上的变化规律;同时,对相应时段的高效益客源的空间分布特征加以分析,为出租车高效益载客热点推荐提供支撑。基于高效益客源的时空分布特征,在对出租车进行高效益载客热点推荐时,首先对出租车GPS数据进行基于效益指标模型的热点分析及热点区域的划分,同时在充分顾及出租车行驶到载客热点区域之后的出租车饱和度及当前的交通拥堵状态的基础上,本文提出了基于出租车效益的推荐度指标模型。该模型以出租车到载客热点的最短行驶时间为基础,计算出租车到达载客热点区域后的出租车供需比及交通运行指数,以此计算推荐度指标来评估该高效益载客热点是否适合推荐给空驶出租车前往,进而形成一套完整的出租车高效益寻客推荐策略方法。最后,本文以深圳市2011年4月18日早高峰的某辆空驶出租车为研究实例,对其进行基于出租车效益指标模型及推荐度指标模型的高效益载客热点推荐,并通过对比分析同一天相同时段内两个高效益载客热点区域的不同推荐结果,同时结合城市居民出行规律及城市功能区结构的实际情况,验证了该推荐策略的有效性。
[Abstract]:With the rapid development of economy and the development of urbanization, people's demand for travel services is increasing. Taxi, as an important part of urban public transportation, plays an important role in people's daily travel. With the development of ITS(Intelligent Transport system technology, many bad phenomena such as high no-load rate of taxis and scrambling for passengers occur frequently because of the unbalanced distribution of passenger space and time, which greatly increases the difficulty of finding taxi passengers and reduces the efficiency of taxi operation. More and more scholars regard taxi passenger seeking mode as the research hotspot. However, there is a general lack of evaluation index of taxi benefit and the study of recommended strategy method, which takes into account the relationship between supply and demand of taxis and traffic conditions. This paper takes the GPS data of 13798 taxis in Shenzhen as the research object, and takes the improvement of taxi benefit as the research goal. Based on the analysis of the factors affecting the efficiency index of taxi passenger seeking, the relationship between supply and demand and the traffic state of the recommended hot spot region of high-benefit passenger transportation are considered, and the method of high-benefit passenger seeking recommendation strategy based on taxi GPS data is put forward. To provide auxiliary decision support for taxi high-benefit passenger seeking, and to improve the unbalanced distribution of taxi resources, at the same time, It is of great significance to give full play to the supplementary role of taxis in urban traffic. By analyzing the index factors that affect the efficiency of taxi passenger seeking, this paper takes the combination of taxi carriage and adjacent no-load as the research object. This paper presents a benefit index model based on the unit time income of taxi carriage and the adjacent no-load seeking time. Based on this model, the data preprocessing and map matching of taxi GPS data are carried out. Using classical mathematical statistics and geostatistical analysis methods, This paper studies the spatial and temporal distribution characteristics of taxi high-benefit passenger source, and indirectly reflects the variation law of high-benefit passenger source with time by comparing and analyzing the changing characteristics of passenger quantity in different periods of working day, non-working day and the same day. At the same time, the spatial distribution characteristics of the high-benefit passenger source in the corresponding period are analyzed to provide the support for the recommendation of the high-benefit passenger carrying hot spot of the taxi. Based on the space-time distribution characteristics of the high-benefit passenger source, when the taxi is recommended to the high-benefit passenger hot spot, Firstly, the hot spot analysis and hot spot area division based on benefit index model are carried out for taxi GPS data. At the same time, based on the full consideration of taxi saturation and current traffic congestion after taxi driving to the hot spot area, In this paper, a recommendation index model based on taxi benefit is proposed. Based on the shortest driving time from taxi to hot spot, the taxi supply and demand ratio and traffic operation index are calculated after the taxi reaches the hot spot. The index of recommendation degree is used to evaluate whether the hot spot of high benefit passenger carrying is suitable for empty taxi, and then a complete set of recommended strategy method of high benefit passenger search for taxi is formed. Finally, Taking an empty taxi in Shenzhen on April 18, 2011, as an example, the hot spot of high benefit passenger carrying is recommended based on taxi benefit index model and recommendation index model. By comparing and analyzing the different recommended results of two high-benefit hot spots in the same day and the same time period, and combining with the actual situation of urban residents' travel rules and the structure of urban functional areas, the effectiveness of the proposed recommendation strategy is verified.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U495;P228.4

【相似文献】