基于行程时间的公交运行可靠度评价及影响因素分析

翁剑成; 赵世昌; 林鹏飞; 孔宁; 钱慧敏

doi:10.3963/j.jssn.1674-4861.2024.06.017

基于行程时间的公交运行可靠度评价及影响因素分析

doi: 10.3963/j.jssn.1674-4861.2024.06.017

1.
北京工业大学交通工程北京市重点实验室北京 100124
2.
北京工业大学计算机学院北京 100124
3.
中交智运有限公司天津 300210
4.
北京市交通运行监测调度中心北京 100161

基金项目:

国家自然科学基金项目 52302381

国家自然科学基金项目 52072011

北京市教育委员会科学研究计划项目 KM202310005025

详细信息

作者简介:
翁剑成（1981—），博士，教授. 研究方向：智能交通、交通出行行为建模等. E-mail：youthweng@bjut.edu.cn

通讯作者:
林鹏飞（1993—），博士，讲师. 研究方向：智能交通、交通大数据等. E-mail：linpengfei@bjut.edu.cn

中图分类号: U491.1+7
计量
- 文章访问数: 44
- HTML全文浏览量: 11
- PDF下载量: 15
- 被引次数: 0
出版历程
- 收稿日期: 2024-04-26
- 网络出版日期: 2025-03-08

Evaluation of Bus Operation Reliability and Analysis of Influencing Factors Based on Travel Time

1.
Beijing University of Technology, Beijing Key Laboratory of Traffic Engineering, Beijing 100124, China
2.
Beijing University of Technology, College of Computer Science, Beijing 100124, China
3.
CCCC Intelligence Transportation Company Limited, Tianjin 300210, China
4.
Beijing Municipal Transportation Operations Coordination Center, Beijing 100161, China

摘要

摘要: 公交运行过程中会受到多种内外部因素干扰，为精准评价公交运行可靠度并量化分析各影响因素。基于公交到站时间数据计算区间行程时间，通过动态阈值和变异系数以及归一化处理，研究了1种可反映不合理延误影响和区间行程时间波动性的公交运行可靠度评价方法，实现不同线路、不同时段公交运行可靠度的横、纵向对比，解决了基于时刻表偏差的公交运行可靠度评价方法不适用于高频服务公交线路的问题。为克服现有研究影响因素考虑单一、以定性分析为主的局限，从站点客流、公交线站属性、道路条件等维度构建8种公交运行可靠度影响因素，利用随机森林模型构建可靠度影响模型，并与支持向量机与反向传播神经网络（back propagation，BP）模型进行精度对比，结合相对重要度和部分依赖图量化分析各影响因素。选取北京市2019年1月9条公交线路的多源公交数据进行实证分析。结果表明：研究提出的评价方法具有良好的有效性，可精准识别早晚高峰期间公交运行的不可靠。采用随机森林构建影响模型的精度最高，相较于支持向量机与BP神经网络分别提升20.38%和49.88%。模型揭示了各个因素的非线性影响机理并确定了有效阈值区间，站点间距、公交区间速度与公交专用道占比是影响公交运行可靠度的关键因素，相对重要度依次为26.9%、25.1%和24.1%。此外，当站点间距在600~800 m之间时，可靠度相较于250 m提升约12.5%；可靠度与公交区间速度呈正相关关系，最高可提升约7%；当公交专用道占比达到60%以上时可靠度显著提升，当占比达到95%时，可靠度提升约6.5%；当途径路段的交叉口数量从1个增加至3个时，可靠度下降约4%；为保证良好的可靠度，公交站台服务的公交线路数不宜超过3条。
- 公交运行可靠度 /
- 区间行程时间 /
- 随机森林模型 /
- 部分依赖图
Abstract: Bus operation is subject to various internal and external factors. To accurately evaluate bus operation reliability and quantitatively analyze the influencing factors. this study calculated the interval travel time based on bus arrival time data. It established a bus operation reliability evaluation method that can reflect the impact of unreasonable delays and the variability of interval travel time by calculating the dynamic threshold probability and coefficient of variation and normalization processing. This method achieves horizontal and vertical comparison of bus operation reliability for different routes and different time periods, solving the problem that the bus operation reliability evaluation method based on schedule deviation is not applicable to high-frequency service bus routes. To address the limitations of existing research, which primarily focuses on single-factor considerations and qualitative analysis, eight influencing factors of bus operation reliability are constructed from perspectives such as station passenger flow, bus route and stop attributes, and road conditions. A Random Forest model is utilized to develop an impact model for bus operation reliability, and its accuracy is compared with that of support vector machine (SVM) and back propagation (BP) Neural Network model. This study used relative importance analysis with partial dependence plots to quantitatively identify key factors and reveal the impact mechanisms. The study uses multi-source bus data from 9 bus routes in Beijing from January 2019 for empirical analysis. The results show that the proposed evaluation method is effective in accurately identifying unreliable bus operations during morning and evening peak hours. The accuracy of the impact model constructed using random forest (RF) is the highest, with improvements of 20.38% and 49.88% compared to SVM and BP Neural Networks, respectively. Key factors influencing reliability include bus stop spacing, bus section speed, and the proportion of dedicated bus lanes, with relative importance values of 26.9%, 25.1%, and 24.1%, respectively. Additionally, the model reveals the nonlinear impact mechanisms of each factor and determines effective threshold intervals. When bus stop spacing is between 600 and 800 m, reliability improves by approximately 12.5% compared to 250 meters. Bus reliability is positively correlated with section speed, with a maximum improvement of around 7%. When the proportion of dedicated bus lanes exceeds 60%, reliability significantly improves, with an increase of about 6.5% when the proportion reaches 95%. Conversely, when the number of signalized intersections along a route increases from 1 to 3, reliability decreases by approximately 4%. To maintain stable reliability, no more than three bus routes should serve the same bus stop.
- bus operation reliability /
- interval travel time /
- random forest model /
- partial dependence plot

HTML全文

图 1 3条线路工作日公交运行可靠度时空分布

Figure 1. Spatio-temporal distribution of bus operation reliability on weekdays for three lines

下载: 全尺寸图片幻灯片

图 2 影响因素的相对重要度

Figure 2. The relative importance of influencing factors

下载: 全尺寸图片幻灯片

图 3 影响因素的部分依赖图

Figure 3. Partial dependency plots of influencing factors

下载: 全尺寸图片幻灯片

表 1 公交站点属性数据样例

Table 1. Samples of bus stop attribute data

线路名称	线路编号	方向	站点编号	站点名称	站点经度/（°）	站点纬度/（°）
1	1	下行	1	老山公交场站	116.233 0	39.914 82
1	1	下行	2	老山南路东口	116.233 9	39.911 73
1	1	下行	3	地铁八宝山站	116.237 3	39.907 34

下载: 导出CSV

表 2 公交到站时间数据样例

Table 2. Samples of bus arrival time data

线路名称	方向	站点名称	车辆编号	到站距离	到站时间	最大站序
430	上行	地坛西门	222 090 918 92	-1	2 019/01/14 08:21:46	23
430	上行	兴化路	222 090 918 92	-1	2 019/01/14 08:25:48	23
430	上行	和平西桥南	222 090 918 92	-1	2 019/01/14 08:29:34	23

下载: 导出CSV

表 3 公交线路属性数据样例

Table 3. Samples of bus route attribute data

线路名称	方向	站点序号	站点名称	距离下一站长度/m
1	下行	1	老山公交场站	352.69
1	下行	2	老山南路东口	658.5
1	下行	3	地铁八宝山站	1 216.39

下载: 导出CSV

表 4 公交乘客智能卡数据样例

Table 4. Samples of bus passenger Smart Card data

智能卡号	线路名称	上车站点编号	下车站点编号	上车时间	下车时间
109 045 539	1	26	20	2 019/1/1 07:34	2 019/1/1 07:48
109 006 789	1	28	17	2 019/1/1 07:10	2 019/1/1 08:00
109 045 541	1	27	20	2 019/1/1 07:32	2 019/1/1 07:50

下载: 导出CSV

表 5 公交运行可靠度影响因素

Table 5. Set of factors affecting bus operation reliability

维度	影响因素	指标释义
站点客流	站点登降量	乘客智能卡数据计算得到的站点上下车人数之和进行赋值/（人·次）
公交线站属性	站点间距	相邻站点之间的距离/m
	站台形式	直线式=1、港湾式=0
	站台服务线路数	站台服务的公交线路数量（1，2，3，……）
道路条件	信号交叉口数量	站点区间内公交运行途径的信号交叉口数量（1，2，3，……）
	公交专用道占比	站点区间内公交专用道长度占总长度的比例
	公交区间速度	公交在站点区间内的运行速度（km/h）
	车道数	公交运行方向车道数量（1，2，3，……）

下载: 导出CSV

表 6 随机森林模型最优参数表

Table 6. Optimal parameter table of random forest model

参数名称	最优参数
最大的弱学习器的个数	70
最大特征数	3
决策树最大深度	19
内部节点再划分所需最小样本数	80
叶子节点最少样本数	10

下载: 导出CSV

表 7 模型精度对比

Table 7. Comparison of accuracy of models

模型	MAE	MSE	NMSE
随机森林模型	0.418	0.292	0.138
支持向量机模型	0.525	0.432	0.198
BP神经网络模型	0.834	0.913	0.341

下载: 导出CSV

参考文献(23)

[1]	WANG P C, HSU Y T, HSU C W. Analysis of waiting time perception of bus passengers provided with mobile service[J]. Transportation Research Part A: Policy and Practice, 2021, 145 (6): 319-336.
[2]	STRATHMAN J G, DUEKER K J, KIMPEL T, et al. Automated bus dispatching, operations control, and service reliability: baseline analysis[J]. Transportation Research Record, 1999, 1666 (1): 28-36. doi: 10.3141/1666-04
[3]	张玉红, 尹小庆, 莫宇迪, 等. 考虑行程时间可靠性的城市公交网络抗毁性研究[J]. 计算机应用研究, 2020, 37(增刊2): 34-36, 40. ZHANG Y H, YIN X Q, MO Y D, et al. Research on the robustness of urban bus networks considering travel time reliability[J]. Application Research of Computers, 2020, 37 (S2): 34-36, 40. (in Chinese)
[4]	童小龙. 公交运行可靠性及其改善策略研究[D]. 南京: 东南大学, 2016. TONG X L. Research on bus operation reliability and its improvement strategies[D]. Nanjing: Southeast University, 2016.
[5]	ZHANG Y, XU H, LU Q C, et al. Travel time reliability analysis considering bus bunching: a case study in xi'an, china[J]. Sustainability, 2022, 14 (23): 1-15.
[6]	RICARD L, DESAULNIERS G, LODI A, et al. Predicting the probability distribution of bus travel time to measure the reliability of public transport services[J]. Transportation Research Part C: Emerging Technologies, 2022, 138: 103619. doi: 10.1016/j.trc.2022.103619
[7]	CHEPURI A, JOSHI S, ARKATKAR S, et al. Development of new reliability measure for bus routes using trajectory data[J]. Transportation Letters, 2020, 12 (6): 363-374. doi: 10.1080/19427867.2019.1595356
[8]	粟海琪, 杨海飞. 降雨对城市公交行程时间可靠性影响分析[J]. 华东交通大学学报, 2022, 39 (3): 80-90. SU H Q, YANG H F. Analysis of the impact of rainfall on the reliability of urban bus travel time[J]. Journal of East China Jiaotong University, 2022, 39 (3): 80-90. (in Chinese)
[9]	严海, 刘润坤. 考虑公交运行可靠性的区间车发车策略[J]. 华南理工大学学报(自然科学版), 2019, 47 (11): 25-32, 43. doi: 10.12141/j.issn.1000-565X.190154 YAN H, LIU R K. Interval bus dispatching strategy considering the reliability of bus operations[J]. Journal of South China University of Technology (Natural Science Edition), 2019, 47 (11): 25-32, 43. (in Chinese) doi: 10.12141/j.issn.1000-565X.190154
[10]	HARSHA M M, MULANGI R H. Impact of side friction on travel time reliability of urban public transit[J]. International Journal of Civil Engineering, 2021, 19 (10): 1221-1237. doi: 10.1007/s40999-021-00622-y
[11]	KAEWUNRUEN S, SRESAKOOLCHAI J, SUN H. Causal analysis of bus travel time reliability in birmingham, uk[J]. Results in Engineering, 2021, 12: 100280. doi: 10.1016/j.rineng.2021.100280
[12]	王玲, 王艳丽, 戚新洲, 等. 基于GPS的常规公交行程时间可靠性因素分析[J]. 武汉理工大学学报(交通科学与工程版), 2022, 46 (1): 1-5. doi: 10.3963/j.issn.2095-3844.2022.01.001 WANG L, WANG Y L, QI X Z, et al. Analysis of factors affecting the reliability of travel time of regular buses based on gps[J]. Journal of Wuhan University of Technology(Transportation Science & Engineering), 2022, 46 (1): 1-5. (in Chinese) doi: 10.3963/j.issn.2095-3844.2022.01.001
[13]	PAUDEL J. Bus ridership and service reliability: the case of public transportation in western massachusetts - sciencedirect[J]. Transport Policy, 2021, 100: 98-107. doi: 10.1016/j.tranpol.2020.10.010
[14]	ANDERSON P, DAGANZO C F. Effect of transit signal priority on bus service reliability[J]. Transportation Research Part B: Methodological, 2020, 132: 2-14. doi: 10.1016/j.trb.2019.01.016
[15]	SOZA-PARRA J, MUOZ J C, RAVEAU S. Factors that affect the evolution of headway variability along an urban bus service[J]. Transportmetrica B, 2021, 9 (1): 479-490.
[16]	SORRATINI J A, LIU R, SINHA S. Assessing bus transport reliability using micro-simulation[J]. Transportation Planning and Technology, 2008, 31 (3): 303-324. doi: 10.1080/03081060802086512
[17]	徐婷, 姜瑞森, 李洪庆, 等. 矩形网格道路多模式公交线网布局优化研究[J]. 重庆交通大学学报(自然科学版), 2019, 38 (10): 108-115. doi: 10.3969/j.issn.1674-0696.2019.10.17 XU T, JIANG R, LI H Q, et al. Research on multi-mode bus network layout optimization for rectangular grid roads[J]. Journal of Chongqing Jiaotong University(Natural Science Edition), 2019, 38 (10): 108-115. (in Chinese) doi: 10.3969/j.issn.1674-0696.2019.10.17
[18]	施卫国, 马万经, 刘兴永. 公交站点排队溢出控制方法[J]. 同济大学学报(自然科学版), 2020, 48 (6): 854-860. SHI W G, MAW J, LIU X Y. Control method for queue overflow at bus stops[J]. Journal of Tongji University(Natural Science Edition), 2020, 48 (6): 854-860. (in Chinese)
[19]	段晓凡, 高良鹏, 简文良, 等. 基于LSTM神经网络的公交线路可靠性评估[J]. 交通工程, 2024, 24 (1): 38-44. DUAN X F, GAO L P, JIAN W L, et al. Reliability evaluation of bus routes based on LSTM neural networks[J]. Journal of Transportation Engineering, 2024, 24(1): 38-44. (in Chinese)
[20]	吴娇蓉, 林清凯, 邓泳淇. 基于公交线路运行稳定性的潜在公交专用道需求识别方法[J]. 吉林大学学报(工学版), 2024, 54 (3): 692-699. WU J R, LIN Q K, DENG Y Q. Identification method of potential demand for bus rapid transit lanes based on bus route operation stability[J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54 (3): 692-699. (in Chinese)
[21]	朱家哲, 赖元文. 道路拥堵程度对公交行程时间可靠性的影响研究[J]. 交通工程, 2017, 17 (3): 47-52. ZHU J Z, LAI Y W. Study on the impact of road congestion on bus travel time reliability[J]. Journal of Transportation Engineering, 2017, 17 (3): 47-52. (in Chinese)
[22]	刘永红, 李进龙, 陈欣, 等. 先进出行信息下事故对行程时间可靠性的影响[J]. 交通运输系统工程与信息, 2018, 18 (3): 36-41. LIU Y H, LI J L, CHEN X, et al. Impact of accidents on travel time reliability under advanced traveler information[J]. Journal of Transportation Systems Engineering and Information Technology, 2018, 18 (3): 36-41. (in Chinese)
[23]	郭继孚, 李先, 王根城. 北京市交通发展年度报告[R]. 北京: 北京交通发展研究院, 2020 GUO J F, LI X, WANG G C. Beijing transportation development annual report[R]. Beijing: Beijing Transportation Development Research Institute, 2020. (in Chinese)