基于多源数据融合的干线公交车辆行程时间预测

交通运输系统工程与信息 ›› 2019, Vol. 19 ›› Issue (4): 124-129.

基于多源数据融合的干线公交车辆行程时间预测

刘迎^{1, 2}，过秀成^*1，周润瑄¹，吕方¹

1. 东南大学交通学院，南京 210096；2. 江苏警官学院治安管理系，南京 210012

收稿日期:2019-01-17 修回日期:2019-04-18 出版日期:2019-08-25 发布日期:2019-08-26
作者简介:刘迎(1988-)，女，江苏徐州人，博士.
基金资助:
江苏省社会科学基金/Social Science Fundation of Jiangsu Province, China(18WTA009).

Travel Time Prediction of Main Transit Line Based on Multi-source Data Fusion

LIU Ying^{1, 2}, GUO Xiu-cheng¹, ZHOU Run-xuan¹, LV Fang¹

1. School of Transportation, Southeast University, Nanjing 210096, China; 2. Department of Security Administration, Jiangsu Police Institute, Nanjing 210012, China

Received:2019-01-17 Revised:2019-04-18 Online:2019-08-25 Published:2019-08-26

摘要/Abstract

摘要：

为提高城市中心区干线公交车辆行程时间的预测精度，在拟合公交车辆行程时间分布特征的基础上，提出基于多源数据的干线公交行程时间预测模型.对RFID及GPS检测器获取的实际数据进行预处理及分布拟合，其中混合高斯分布函数适用于单路段拟合，对数正态分布适用于多路段的拟合.采用皮尔逊相关性系数对影响行程时间的因素进行相关性分析，其中上游路段前2 个时间窗的平均行程时间的影响最大.分别采用ARIMA、改进的SVM模型对行程时间进行预测，其中改进的SVM模型的平均绝对百分比误差为6.26%，优于ARIMA模型的11.69%，更适用于短距离交叉口间的公交车辆行程时间预测.

关键词: 城市交通, 干线公交, 数据融合, 行程时间预测

Abstract:

In order to improve the travel time prediction accuracy of the main transit line in the city center, based on the analysis of the distribution characteristics of the bus travel time, the method of forecasting the transit time based on multi- source data is proposed. The actual data obtained by radio frequency identification and global positioning equipment are preprocessed, and the fitting analysis is carried out by mathematical statistics model. The mixed Gaussian distribution function is suitable for single-segment fitting, and the lognormal distribution is suitable for multi- segment fitting. The Pearson correlation coefficient is used to sort the time series and spatial factors affecting the travel time. The average travel time of the first two time windows of the upstream road segment has the greatest impact. The travel time is predicted by using ARIMA and SVM models, respectively. The mean absolute percentage error of the ARIMA model is 11.69%, and the mean absolute percentage error of the support vector machine model is 6.26%.

Key words: urban traffic, main transit line, data fusion, travel time prediction

中图分类号:

U491

刘迎，过秀成，周润瑄，吕方. 基于多源数据融合的干线公交车辆行程时间预测[J]. 交通运输系统工程与信息, 2019, 19(4): 124-129.

LIU Ying, GUO Xiu-cheng, ZHOU Run-xuan, LV Fang. Travel Time Prediction of Main Transit Line Based on Multi-source Data Fusion[J]. Journal of Transportation Systems Engineering and Information Technology, 2019, 19(4): 124-129.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/abstract/abstract19879.shtml

http://www.tseit.org.cn/CN/Y2019/V19/I4/124

参考文献

[1] 李克平. 城市公共交通优先的管理与控制[J]. 交通与运输, 2017, 9(5): 1-3. [LI K P. Management and control of urban public transportation priority[J]. Traffic & Transportation, 2017, 9(5): 1-3.]

[2] BERREBI S, HANS E, CHIABAUT N, et al. Comparing bus holding methods with and aithout real- time predictions[J]. Transportation Research Part C, 2018 (87): 197-211.

[3] MA Z L, KOUTSOPOULOS H, FERREIRA L, et al. Estimation of trip travel time distribution using a generalized markov chain approach[J]. Transportation Research Part C, 2017(74): 1-21.

[4] 姚志洪, 蒋阳升, 赵斌, 等. 行程时间服从混合高斯分布的车队离散模型[J]. 交通运输系统工程与信息, 2017, 17(2): 101- 108, 129. [YAO Z H, JIANG Y S, ZHAO B, et al. Platoon dispersion model based on mixed gaussian distribution of travel time[J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(2): 101-108, 129.]

[5] 龚越, 罗小芹, 王殿海, 等. 基于梯度提升回归树的城市道路行程时间预测[J]. 浙江大学学报(工学版), 2018, 52(3): 46-53. [GONG Y, LUO X Q, WANG D H, et al. Urban travel time prediction based on gradient boosting regression trees[J]. Journal of Zhejiang University (Engineering Science), 2018, 52(3): 46-53.]

[6] 杨帆, 杨晓光, 云美萍. 城市道路间断流行程时间双峰分布[J]. 同济大学学报(自然科学版), 2014, 42(2): 97- 103. [YANG F, YANG X G, YUN M P. Urban road interrupted flow travel time bimodal distribution[J]. Journal of Tongji University(Natural Science), 2014, 42 (2): 97-103.]

[7] TANG K, CHEN S Y, LIU Z Y. A tensor-based bayesian probabilistic model for citywide personalized travel time estimation[J]. Transportation Research Part C, 2018(90): 260-280.

[8] 杨敏, 丁剑, 王炜. 基于ARIMA-SVM模型的快速公交停站时间组合预测方法[J]. 东南大学学报(自然科学版), 2016, 46(3): 651-656. [YANG M, DING J, WANG W. Hybrid dwell time prediction method for bus rapid transit based on ARIMA-SVM model[J]. Journal of Southeast University (Natural Science Edition), 2016, 46 (3): 651-656.]

[9] YU Z Y, WOOD J, GAYAH V. Using survival models to estimate bus travel times and associated uncertainties [J]. Transportation Research Part C, 2017(74): 366-382.

[10] AIDAN O, FRANCISCO P C, ZHAO J H, et al. Uncertainty in bus arrival time predictions: Treating heteroscedasticity with a metamodel approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 17(11): 3286-3296.

[11] CHANG H, PARK D, LEE S, et al. Dynamic multiinterval bus travel time prediction using bus transit data [J]. Transportmetrica, 2010, 6(1): 19-38.

[12] ODED C. An agent-based approach for modeling realtime travel information in transit systems[J]. Procedia Computer Science, 2014(32): 744-749.

[13] YU B, YANG Z Z, CHEN K, et al. Hybrid model for prediction of bus arrival times at next station[J]. Journal of Advanced Transportation, 2010, 44(3): 193-204.

[14] YU B, LAM W H, TAM M L. Bus arrival time prediction at bus stop with multiple routes[J]. Transportation Research Part C: Emerging Technologies, 2011, 19(6): 1157-1170.

[15] 汪林. 基于预测的快速公交信号优先设计及效果仿真[J]. 公路交通科技, 2017, 34(6): 129- 135. [WANG L. Signal priority design for bus rapid transit based on prediction method and effect simulation[J]. Journal of Highway and Transportation Research and Development, 2017, 34(6): 129-135.]

[16] BIE Y, WANG D H, QI H S. Prediction model of bus arrival time at signalized intersection using GPS data[J]. Journal of Transportation Engineering-Asce, 2012, 138 (1): 12-20.

[17] RAHMAN M, WIRASINGHE S, KATTAN L. Analysis of bus travel time distributions for varying horizons and real-time applications[J]. Transportation Research Part C, 2018(86): 453-466.