基于乘客群体出行时间间隔的标度律研究

doi:10.16088/j.issn.1001-6600.2020.01.001

广西师范大学学报（自然科学版） ›› 2020, Vol. 38 ›› Issue (1): 1-9.doi: 10.16088/j.issn.1001-6600.2020.01.001

• • 下一篇

基于乘客群体出行时间间隔的标度律研究

李飞羽, 翁小雄^*, 姚树申

华南理工大学土木与交通学院,广东广州510640

收稿日期:2019-01-08 出版日期:2020-01-25 发布日期:2020-01-15
通讯作者: 翁小雄(1958—),女,浙江杭州人,华南理工大学教授,博导。E-mail:ctxxweng@qq.com
基金资助:
国家自然科学基金 (51578247);广东省交通运输厅科技项目 (科技-2015-02-076)

Research on the Scaling Law Based on the Travel Time Interval of Passengers’ Group

LI Feiyu, WENG Xiaoxiong^*, YAO Shushen

School of Civil Engineering and Transportation, South China University of Technology,Guangzhou Guangdong 510640, China

Received:2019-01-08 Online:2020-01-25 Published:2020-01-15

摘要/Abstract

摘要： 为了深入研究乘客出行时间间隔的标度律及其形成的机制,本文基于粒子群算法对乘客群体出行时间间隔的概率分布进行了拟合,发现了群体时间间隔不服从统一的幂律分布。基于拟合得到的结果,分别从阵发性、乘客出行时间规律以及出行目的三方面进行了分析。分析结果表明,78.2%的时间间隔聚集在14 h范围内,21.8%的时间间隔大于14 h,从而使得两个范围对应着不同的幂指数。群体时间间隔的标准差值大于平均值,导致分布出现了胖尾现象。群体时间间隔聚集了大量随机出行乘客的时间间隔,使得群体的时间间隔偏离了泊松分布。不同目的出行的乘客具有不同的时间间隔范围,从而导致了群体时间间隔的分布不均匀。在此结果上挖掘出乘客出行时间间隔主要集中在14 h内,从而为交通管理部门预测乘客出行时间提供了一种新的思路。

关键词: 乘客, 时间间隔, 粒子群算法, 幂律分布, 胖尾现象, 泊松分布, 出行时间, 概率分布

Abstract: In order to research the scaling law of the passenger travel time interval and the formation mechanism for the scaling law, the particle swarm optimization algorithm is used to fit the probability distribution of the travel time intervals of the passengers’ group. It is found that the time intervals of the group do not obey the unified power law distribution. Based on the results obtained by the fitting, the analysis is made from three aspects: the burstiness, the regularity of the passenger’s travel time and the purpose of travel. The results show that the 78.2% of time interval is concentrated within the 14 hours and the 21.8% of time intervals is greater than 14 hours, which makes the two ranges correspond to different power exponents. The value of the standard deviation of the time intervals is greater than its average value, resulting in a fat tail in the distribution. A large number of time intervals of random travel passengers causes the time intervals of passengers’group deviate from the poisson distribution. Passengers traveling for different purposes have different time interval ranges, resulting in uneven time intervals of the passenger’ group. These results indicate that passenger travel time intervals are mainly concentrated within 14 hours, which provides a new way for the traffic management department to predict the passenger travel time.

Key words: passenger, time interval, particle swarm optimization, power law distribution, fat tail phenomenon, poisson distribution, travel time, probability distribution

中图分类号:

U491

李飞羽, 翁小雄, 姚树申. 基于乘客群体出行时间间隔的标度律研究[J]. 广西师范大学学报（自然科学版）, 2020, 38(1): 1-9.

LI Feiyu, WENG Xiaoxiong, YAO Shushen. Research on the Scaling Law Based on the Travel Time Interval of Passengers’ Group[J]. Journal of Guangxi Normal University(Natural Science Edition), 2020, 38(1): 1-9.

参考文献

[1] 吕攀龙,翁小雄,彭新建.基于差分进化算法SVM的公交通勤乘客识别[J].广西师范大学学报(自然科学版),2019,37(1):106-114.
[2] 肖梅,王仲豪,颜建强,等.基于西安城市公交IC卡数据的乘客上车耗时影响因素分析[J].重庆理工大学学报(自然科学),2019,33(1):136-142.
[3] BARABASI A. The origin of bursts and heavy tails in human dynamics[J]. Nature, 2005, 435(7039): 207-211.
[4] OLIVEIRA J G, BARABASI A L. Darwin and Einstein correspondence patterns [J]. Nature,2005,437:1251.
[5] VÁZQUEZ A, OLIVEIRA J G, DEZS Z, et al. Modeling bursts and heavy tails in human dynamics[J]. Phys Rev E, 2006, 73(3): 036127.
[6] ZHAO Z D, ZHOU T.Empirical analysis of online human dynamics[J].Physica A: Statistical Mechanics and its Applications, 2012, 391(11):3308-3315.
[7] 徐宇.突发公共事件中的微博公众参与行为研究[D].成都:电子科技大学,2017.
[8] 樊超, 宗利永. MOOC在线学习行为的人类动力学分析[J]. 开放教育研究, 2016, 22(2): 53-58.
[9] 姜春昕.社会-网络交融空间中人类行为模式与社交结构分析技术[D].天津:天津大学,2018.
[10]李瑾颉, 吴联仁, 齐佳音, 等. 基于人类动力学的在线社交网络信息传播研究[J]. 电子与信息报, 2017, 39(4): 785-793.
[11]徐达.基于行为动力学的微博用户行为分析[D].北京:中国人民公安大学,2018.
[12]张杰斌,秦红.微博用户的时间行为研究[J].计算机科学,2016,43(6):24-27.
[13]崔晓萌. 基于浮动车数据的居民出行行为的动力学模型及特征分析[D].哈尔滨:哈尔滨工业大学,2015.
[14]沈洪洲, 史俊鹏. 基于人类动力学的社会化问答社区优秀贡献者行为研究:以“知乎”为例[J]. 情报科学, 2019, 37(5): 85-91.
[15]YANG T, FENG X, WU Y, et al. Human dynamics in repurchase behavior based on comments mining[J]. Physica A: Statistical Mechanics and Its Applications, 2018, 502: 563-569.
[16]YAN D C , WEI Z W , HAN X P , et al. Empirical analysis on the human dynamics of blogging behavior on GitHub[J]. Physica A: Statistical Mechanics and its Applications, 2016, 465: 775-781.
[17]张艳丰.在线用户评论行为时间序列关联特征规律研究[D].长春:吉林大学,2018.
[18]LI K, LV T, SHEN H, et al. An empirical analysis on the behavioral differentia of the “Elite-Civilian” users in Sina microblog[J]. Physica A: Statistical Mechanics and Its Applications, 2019, 539: 122974.
[19]ZHOU C, CHEN R, JIANG S, et al. Human dynamics in near-miss accidents resulting from unsafe behavior of construction workers[J]. Physica A: Statistical Mechanics and Its Applications, 2019, 530: 121495.
[20]杨光. 快速公交乘客出行的时空规律特征研究[J]. 物流技术, 2017, 36(3): 97-100.
[21]GOH K I, BARABÁSI A L. Burstiness and memory in complex systems[J]. Europhys Lett, 2008, 81(4): 48002.
[22]余胜威, MATLAB优化算法案例分析与应用[M].北京:清华大学出版社,2014,155-179.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于乘客群体出行时间间隔的标度律研究

Research on the Scaling Law Based on the Travel Time Interval of Passengers’ Group

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

Metrics

本文评价

推荐阅读 10

[1]	吕攀龙, 翁小雄, 彭新建. 基于差分进化算法SVM的公交通勤乘客识别[J]. 广西师范大学学报（自然科学版）, 2019, 37(1): 106-114.
[2]	胡郁葱, 陈栩, 罗嘉陵. 多起终点多车型混载的定制公交线路规划模型[J]. 广西师范大学学报（自然科学版）, 2018, 36(4): 1-11.
[3]	王剑, 黄植功, 许金海. 基于优化EKF的永磁同步电机转速估计[J]. 广西师范大学学报（自然科学版）, 2014, 32(4): 11-17.
[4]	严晓明, 郑之. 基于混合仿生算法的SVM参数优化[J]. 广西师范大学学报（自然科学版）, 2011, 29(2): 114-118.