出租车高效益行程特征

doi:10.16097/j.cnki.1009-6744.2023.04.031

交通运输系统工程与信息 ›› 2023, Vol. 23 ›› Issue (4): 307-314.DOI: 10.16097/j.cnki.1009-6744.2023.04.031

• 工程应用与案例分析 • 上一篇

出租车高效益行程特征

年光跃¹，潘海啸^*1，孙健²

1. 同济大学，建筑与城市规划学院，上海 200092；2. 长安大学，未来交通学院，西安 710021

收稿日期:2023-04-02 修回日期:2023-05-28 接受日期:2023-06-19 出版日期:2023-08-25 发布日期:2023-08-22
作者简介:年光跃(1986- )，男，安徽蚌埠人，博士后
基金资助:
国家自然科学基金区域创新发展联合基金(U20A20330)

Exploring High-income Trip Characteristics of Taxis

NIAN Guang-yue¹, PAN Hai-xiao^*1, SUN Jian²

1. College of Architecture and Urban Planning, Tongji University, Shanghai 200092, China; 2. School of Future Transportation, Chang'an University, Xi'an 710021, China

Received:2023-04-02 Revised:2023-05-28 Accepted:2023-06-19 Online:2023-08-25 Published:2023-08-22
Supported by:
National Natural Science Foundation of China (NSFC) Regional Innovation Development Joint Fund Project (U20A20330)

摘要/Abstract

摘要： 为提高出租车营运效率和服务水平，以重庆为案例，研究出租车行程特征与出租车驾驶员单位时间收入的关联机理。首先，基于关联的出租车营运数据和轨迹数据构建行程特征指标；其次，构建随机森林预测模型分析行程特征对单位时间收入的相对重要性及显著性；然后，构建多分类Logistic回归模型分析行程特征对单位时间收入的量化影响。研究结果表明：行程特征指标可用于预测驾驶员单位时间收入且准确性较好，送客速度、寻客时长、长单数量对平均单位时间收入预测的相对重要性最大；寻客行程迂回度、送客行程迂回度、寻客里程的增加会显著促使普通效益驾驶员降为低效益驾驶员，而寻客行程迂回度、寻客区域偏好度、寻客里程的减少及送客速度的增加会显著促使普通效益驾驶员上升为高效益驾驶员。高效益驾驶员具有寻客积极主动、不偏好特定区域、倾向预期通行速度快的短路径、有长单偏好但不刻意追求等特征。本文弥补了先前相关研究在行程特征与效益的关联性构建，多样化行程特征对单位时间收入的表征与预测，高效益出租车驾驶员营运特征勾勒等方面的不足；可为城市交通管理、出租车数量管控及运价调整等提供理论依据和技术支撑。

关键词: 城市交通, 单位时间收入, 随机森林, 多分类逻辑斯蒂回归, 出租车驾驶员, 行程指标

Abstract: To improve the operational efficiency and service level of taxis, the correlation mechanism between taxi trip characteristics and taxi drivers' income per unit time (IPUT) is studied based on the central area of Chongqing. A random forest prediction model is constructed to analyze the relative importance and significance of trip characteristics on IPUT. The results show that trip characteristics can be used to predict drivers' IPUT with good accuracy, and the relative importance of delivery speed, search time, and the number of long orders is the greatest in predicting the average IPUT. The increase in search trip detour, delivery trip detour, and search mileage significantly increases the probability that the middle-income drivers fall to the low-income drivers, while the decrease in search trip detour, search area preference, search mileage, and delivery speed significantly increase the probability that the middle-income drivers rise to the high-income drivers. High-income drivers have the characteristics of being proactive in searching for passengers, not preferring specific areas, tending to anticipate short routes with high travel speeds, and favoring long orders but not deliberately pursuing them. This study bridges the gaps in previous research on the construction of correlations between trip characteristics and benefits, the characterization and prediction of diverse trip characteristics on IPUT, and the outlining of operating characteristics of high-income taxi drivers. The study may provide theoretical references and technical support for urban traffic management, taxi quantity regulation, and fare adjustment.

Key words: urban traffic, wage rate, random forest, multinomial Logistic regression, taxi driver, trip indicator

中图分类号:

U121

年光跃, 潘海啸, 孙健. 出租车高效益行程特征[J]. 交通运输系统工程与信息, 2023, 23(4): 307-314.

NIAN Guang-yue, PAN Hai-xiao, SUN Jian. Exploring High-income Trip Characteristics of Taxis[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(4): 307-314.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/10.16097/j.cnki.1009-6744.2023.04.031

http://www.tseit.org.cn/CN/Y2023/V23/I4/307

参考文献

[1] YANG H, FUNG C S, WONG K I, et al. Nonlinear pricing of taxi services[J]. Transportation Research Part A: Policy and Practice, 2010, 44(5): 337-348. DOI: 10.1016/j.tra.2010.03.004.

[2] ZHANG K, CHEN Y, NIE Y (Marco). Hunting image: Taxi search strategy recognition using Sparse Subspace Clustering[J]. Transportation Research Part C: Emerging Technologies, 2019, 109: 250-266. DOI: 10.1016/j. trc.2019.10.015.

[3] LIANG Y, ZHAO Z, ZHANG X. Modeling taxi cruising time based on multi-source data: A case study in Shanghai[J]. Transportation, 2022, 11(9): 1-30. DOI: 10.1007/s11116-022-10348-y.

[4] ZHANG D, SUN L, LI B, et al. Understanding taxi service strategies from taxi GPS traces[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(1): 123-135. DOI:10.1109/TITS.2014.2328231.

[5] JI Y, DU Y, LIU Y, et al. Empirical behavioral study of airport-serving taxi drivers using automatic vehicle location data[J]. Journal of Urban Planning and Development, 2017, 143(1): 4016026. DOI:10.1061/ (ASCE)UP.1943-5444.0000353.

[6] YAZICI M A, KAMGA C, SINGHAL A. Modeling taxi drivers' decisions for improving airport ground access: John F. Kennedy airport case[J]. Transportation Research Part A: Policy and Practice, 2016, 91: 48-60. DOI:10.1016/j.tra.2016.06.004.

[7] QIN G, LI T, YU B, et al. Mining factors affecting taxi drivers' incomes using GPS trajectories[J]. Transportation Research Part C: Emerging Technologies, 2017, 79: 103- 118. DOI:10.1016/j.trc.2017.03.013.

[8] 周悦, 江欣国, 付川云, 等. 基于混合Logit模型的出租车超速者运营因素分析[J]. 交通运输系统工程与信息, 2021, 21(3): 229-236. DOI: 10.16097/ j.cnki.1009- 6744.2021.03.029. [ZHOU Y, JIANG X G, FU C Y, et al. Operational factors analysis for taxi speeders using mixed Logit model[J]. Journal of Transportation Systems Engineering and Information Technology, 2021, 21(3): 229-236. DOI: 10.16097/ j. cnki.1009-6744.2021.03.029.]

[9] DEMISSIE M G, KATTAN L, PHITHAKKITNUKOON S, et al. Modeling location choice of taxi drivers for passenger pickup using GPS data[J]. IEEE Intelligent Transportation Systems Magazine, 2021, 13(1): 70- 90. DOI:10.1109/MITS.2020.3014099.

[10] 鞠炜奇, 杨家文, 林雄斌. 城市出租车空载率时空特征及其影响因素研究: 以深圳市为例[J]. 规划师, 2015, 31(S2): 257-262. DOI: 10.3969/j.issn.1006-0022.2015. z2.050. [JU W Q, YANG J W, LIN X B. Research on the space-time characteristics and influence factors of taxi empty-loading ratio: Shenzhen example[J]. Planners, 2015, 31(S2): 257-262. DOI: 10. 3969/ j.issn. 1006- 0022.2015.z2.050.]

[11] GAO Y, XU P, LU L, et al. Visualization of taxi drivers' income and mobility intelligence[C]//Advances in Visual Computing-8th International Symposium (ISVC 2012). Springer, Berlin, Heidelberg, 2012: 275-284. DOI: 10.1007/978-3-642-33191-6_27.

[12] HU B, XIA X, SHEN X, et al. Analyses of the imbalance of urban taxis' high-quality customers based on Didi trajectory data[J]. Journal of Advanced Transportation, 2019. DOI:10.1155/2019/3689389.

[13] LIU L, WU C, ZHANG H, et al. Research on taxi driverss' passenger hotspot selecting patterns based on GPS data: A case study in Wuhan[C]//2017 4th International Conference on Transportation Information and Safety(ICTIS). IEEE, Piscataway, NJ, 2017: 432- 441. DOI:10.1109/ICTIS.2017.8047802.

[14] ZONG F, ZHANG H, ZHU X, et al. Understanding taxi driverss' multi-day cruising patterns[J]. Promet-Traffic & Transportation, 2015, 27(6): 467- 476. DOI: 10.7307/ptt.v27i6.1641.

[15] ZHENG Z, RASOULI S, TIMMERMANS H. Modeling taxi driver search behavior under uncertainty[J]. Travel Behaviour and Society, 2021, 22: 207-218. DOI:10.1016/ j.tbs.2020.09.008.

[16] LIU L, ANDRIS C, RATTI C. Uncovering cabdrivers' behavior patterns from their digital traces[J]. Computers, Environment and Urban Systems, 2010, 34(6): 541-548. DOI:10.1016/j.compenvurbsys.2010.07.004.

[17] 邱端昇, 邬群勇, 刘萌, 等. 一种基于出租车轨迹数据的优质客源评价模型及实证研究[J]. 福州大学学报(自然科学版), 2018, 46(2): 199-203. DOI: 10.7631/ issn.1000-2243.16423. [QIU D S, WU Q Y, LIU M, et al. A model of high quality passenger based on taxi trajectory and case verification[J]. Journal of Fuzhou University(Natural Science), 2018, 46(2): 199-203. DOI: 10.7631/issn.1000-2243.16423.]

[18] 刘丽, 张丰, 杜震洪, 等. 基于深圳市出租车轨迹数据的高效益寻客策略研究[J]. 浙江大学学报(理学版), 2018, 45(1): 82-91. DOI: 10. 3785/ j.issn. 1008- 9497.2018.01.013. [LIU L, ZHANG F, DU Z H, et al. The analysis of high profitable strategy for seeking passengers based on taxi GPS trajectory data of Shenzhen city[J]. Journal of Zhejiang University(Science Edition), 2018, 45(1): 82-91. DOI: 10.3785/j.issn.1008- 9497.2018.01.013.]

[19] BREIMAN L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32. DOI:10.1023/A:1010933404324.

[20] CRAWFORD V P, MENG J. New York City cab drivers' labor supply revisited: Reference-dependent preferences with rational-expectations targets for hours and income [J]. American Economic Review, 2011, 101(5): 1912- 1932. DOI:10.1257/aer.101.5.1912.

出租车高效益行程特征

Exploring High-income Trip Characteristics of Taxis

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	杨洋, 王雪纯, 袁振洲, 陈进杰, 那艳玲. 基于碳回收期理论的城市轨道交通碳减排效应测算[J]. 交通运输系统工程与信息, 2023, 23(5): 1-11.
[2]	周涛, 李毅军, 孙琴梅, 任瀚堃, 刘怡, 张振豪. 大数据驱动下的山地城市道路条件对车辆碳排放影响研究[J]. 交通运输系统工程与信息, 2023, 23(5): 172-183.
[3]	李雪岩, 李海洋, 张汉坤. 考虑减排的公交线网分时差异化票价动态优化[J]. 交通运输系统工程与信息, 2023, 23(5): 12-23.
[4]	周洋, 杨超, 郭唐仪. 基于手机GPS数据的个体出行链提取方法研究[J]. 交通运输系统工程与信息, 2023, 23(5): 45-54.
[5]	常玉林, 蔡宇航, 孙超, 王建. 异质需求下响应型接驳公交车速引导与路径规划协同研究[J]. 交通运输系统工程与信息, 2023, 23(5): 83-95.
[6]	李光春, 聂磊. 基于改进自适应大邻域算法的公交线网规划方法[J]. 交通运输系统工程与信息, 2023, 23(5): 96-103.
[7]	徐士伟, 苏业辉, 李慧文, 王帅, 江欣国, 马剑. 基于熵权法的枢纽内公交站场布局评价研究[J]. 交通运输系统工程与信息, 2023, 23(5): 104-112.
[8]	姚志刚, 卢致远. 联合重复性与周期性的公交乘客个体出行规律分类[J]. 交通运输系统工程与信息, 2023, 23(5): 113-119.
[9]	郑乐, 高良鹏, 陈学武, 宋波, 丁磊. 地铁-公交加权复合网络关键站点识别及鲁棒性研究[J]. 交通运输系统工程与信息, 2023, 23(5): 120-129.
[10]	马健霄, 赵飞燕, 尹超英, 汤文蕴. 建成环境和出租车需求对网约车出行需求影响的时空间分异模式[J]. 交通运输系统工程与信息, 2023, 23(5): 136-145.
[11]	周成栋, 宋菲, 赵小梅. 考虑共享汽车和区域换乘的多模式双动态演化模型[J]. 交通运输系统工程与信息, 2023, 23(5): 146-154.
[12]	任婧璇, 常孝亭, 巫威眺, 靳文舟. 考虑候选站点和全服务过程的需求响应接驳公交调度[J]. 交通运输系统工程与信息, 2023, 23(5): 202-214.
[13]	于丁原, 姚恩建, 刘莎莎, 李思慧, 郭东博, 刘唯伊. 运营突发事件下城市轨道交通诱导信息发布策略研究[J]. 交通运输系统工程与信息, 2023, 23(5): 227-237.
[14]	许得杰, 潘星, 巩亮, 胡晨皓, 王雪鑫. 基于列车开行方案的城市轨道交通多交路客流分配方法[J]. 交通运输系统工程与信息, 2023, 23(5): 238-246.
[15]	张惠茹, 豆飞, 魏运, 刘洁, 宁尧. 高峰时段郊区线考虑列车容量约束的停站方案优化方法研究[J]. 交通运输系统工程与信息, 2023, 23(5): 247-257.