地球信息科学学报 ›› 2017, Vol. 19 ›› Issue (2): 167-175.doi: 10.3724/SP.J.1047.2017.00167

• 地理空间分析综合应用 • 上一篇    下一篇

出租车交接班行为识别与时空分布研究

唐炉亮(), 段倩*(), 阚子涵, 李清泉   

  1. 武汉大学 测绘遥感信息工程国家重点实验室,武汉 430079
  • 收稿日期:2016-06-16 修回日期:2016-08-05 出版日期:2017-02-28 发布日期:2017-02-17
  • 通讯作者: 段倩 E-mail:tll@whu.edu.cn;d.q@whu.edu.cn
  • 作者简介:

    作者简介:唐炉亮(1973-),男,博士生导师,珞珈特聘教授,研究方向为GIS-T、时空数据获取与分析等。E-mail: tll@whu.edu.cn

  • 基金资助:
    国家自然科学基金项目(41571430、41271442、40801155)

Study on Identification and Space-time Distribution Analysis of Taxi Shift Behavior

TANG Luliang(), DUAN Qian*(), KAN Zihan, LI Qingquan   

  1. State Key Laboratory of Information and Engineering in Surveying, Mapping and Remote Sensing,Wuhan University, Wuhan 430079, China
  • Received:2016-06-16 Revised:2016-08-05 Online:2017-02-28 Published:2017-02-17
  • Contact: DUAN Qian E-mail:tll@whu.edu.cn;d.q@whu.edu.cn

摘要:

在出租车交接班时间段内经常发生打车难,甚至空车拒载的现象,研究出租车交接班行为的时空分布特征不仅能提高出租车运行效率,同时能缓解出租车供需矛盾,方便公众出行。本文采用武汉市出租车GPS轨迹数据,建立了出租车交接班时空序列模式,提出了交接班时空序列模式挖掘与识别方法,分析了交接班事件的时空分布特征,并以车辆区域覆盖强度、区域覆盖密度为指标对交接班停靠点城市资源配置进行了评估。研究结果表明:武汉市出租车交接班集中发生在凌晨1:00-4:00与下午16:00-17:00,下午交接班高峰与晚交通高峰时段有部分重叠,交接班地点比较均匀地遍布中心城区(青山区除外,此区域开发程度较低),武昌区交接班强度最大,江汉区交接班密度最大。此外,结合武汉市2012年出台禁止出租车司机在晚高峰交通时段交接班的规定,探测发现6.5%左右的司机仍存在严重违规交接班行为。

关键词: 交接班, GPS时空轨迹, 时空序列模式, 时空分布

Abstract:

In the period of taxis taking shift, many city dwellers find it hard to get a taxi or even get turned away in front of vacant ones. Analyzing the spatial and temporal distribution of taxi shift change behavior can improve taxis efficiency, relieve the contradictions of supply and demand of taxi, and facilitate public travelling. This study establishes space-time sequence of taxi shift behavior and puts forward a method to mine and identify taxi shift sequence from GPS trajectory. Furthermore, we adopt taxis′ GPS trajectory data in Wuhan to perform our method. Based on the identified shift events, we analyze the space-time distribution of taxis′ shift behavior and evaluate parking resource allocation for taxis′ shift change by using intensity and density as indicators. The results show that: taxis′ shift behavior in Wuhan peaks in period of 1:00 to 4:00 and 16:00 to 17:00 and the latter peak in the afternoon partially overlaps with evening rush hours, which can introduce the difficulty in taking a taxi; the parking location for taxis taking shift is relatively uniformly distributed throughout the downtown of the city except Qingshan District, which is less prosperous than the others; Wuchang District has the strongest shift intensity and Jianghan District, whose area is the smallest among all the administrative districts, has the greatest shift density. In addition, considering the regulation issued by Wuhan traffic administration in 2012, which prohibits taxi drivers from taking shift in the evening rush hours, it is revealed that about 6.5% of the drivers have serious illegal shift behavior.

Key words: shift, GPS space-time trajectory, space-time sequential pattern, space-time distribution