快速公交乘客出行的时空规律特征研究

发布时间：2018-11-03 07:05

【摘要】：人类移动行为的定量化研究,包括时间特性和空间特性,.对于理解许多由人群交互所产生的传播现象至关重要,并且在空间位置预测、信息推荐等领域有很大帮助。近年来大数据技术和城市公共信息管理系统的兴起,使得获取城市人群大规模行为数据成为可能。本文借助成都市快速公交智能卡数据,定义并定量研究了乘客出行的时空规律特征,并且采用信息熵定义了乘客出行的空间熵和时间熵。研究表明,快速公交乘客出行在时间和空间上都存在异质性。在时间上,乘客日常出行具有明显的早晚高峰现象,周末出行量比工作日出行量少,早晚高峰现象也变得不明显,90%的乘客的出行在途时间不超过30分钟,乘客的连续出行多发生在1.5小时左右的短时间、8.5至13.5小时的通勤出行以及以24小时的倍数依次衰退的周期3个时间段。在空间上,大部分乘客出行具有明显的空间偏好,一些站点的访问次数很多,也有站点的访问很稀疏。而对于同一站点,进站和出站频率往往并不对称。对于大多数乘客,受到物理因素的限制,只访问过系统中的少数站点。而乘客出行的移动距离分布,既不服从指数分布,也不服从幂函数分布,而是具有若干离散的峰值,这可能与站点间的距离分布以及线路较为单一有关。快速公交出行的时空规律特征的回归性分析表明,乘客出行的时间熵与出行绝对时间差存在正相关关系,而空间熵随着乘客访问站点个数的增加而增大,随着乘客出行频率的增加而减小。时空熵各自的分布表明,乘客出行在空间上受到的限制更大,而在时间上表现出很强的随机性和离散性。时间熵和空间熵对于乘客出行的首次刷卡进站时间、出行频率以及访问站点的个数有不同程度的影响。
[Abstract]:Quantitative study of human mobility, including temporal and spatial characteristics. It is very important to understand many communication phenomena caused by crowd interaction, and it is helpful in spatial location prediction, information recommendation and so on. In recent years, with the rise of big data technology and urban public information management system, it is possible to obtain mass behavior data of urban population. Based on the smart card data of Chengdu bus Rapid Transit (BRT), this paper defines and quantitatively studies the spatial and temporal characteristics of passenger travel, and uses information entropy to define the spatial entropy and time entropy of passenger travel. The research shows that there is heterogeneity in time and space of bus rapid transit passenger travel. In terms of time, the daily travel of passengers has obvious morning and evening rush phenomenon, weekend travel volume is less than weekday travel volume, morning and evening rush time phenomenon become not obvious, 90% of passengers travel time is less than 30 minutes, Passengers travel continuously in about 1.5 hours of short time, 8.5 to 13.5 hours of commuting and 24 hours of decline in turn of three periods of time. In space, most passengers travel with obvious spatial preference, some sites have a lot of visits, and others have very sparse access. For the same station, the frequency of incoming and outgoing stations is often asymmetric. For most passengers, physical constraints have limited access to only a few sites in the system. However, the travel distance distribution of passengers is neither exponential nor power function, but has some discrete peaks, which may be related to the distance distribution between stations and the single line. The regression analysis of the spatial and temporal characteristics of bus rapid transit travel shows that there is a positive correlation between the travel time entropy and the absolute travel time difference, and the spatial entropy increases with the increase of the number of passenger visits. As the frequency of passenger travel increases, it decreases. The spatial and temporal entropy distribution shows that the travel of passengers is more restricted in space, but it shows strong randomness and discreteness in time. The time entropy and the space entropy have different effects on the first time of swiping in the station, the frequency of the trip and the number of sites visited.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491

【参考文献】