基于LightGBM算法的公交行程时间预测
|
摘要
在城市公交网络运行中,公交车的站点间行程时间会受到道路和环境条件的影响.本文对公交车运行过程中的车辆速度特征、道路特征及天气特征等进行了分析.建立了基于特征的LightGBM (Light Gradient Boosting Machine)公交行程时间预测模型,通过调整LightGBM算法中的相关参数,以分配各个影响特征和因素的权重大小.然后利用天津市某条公交线路24天的公交车GPS数据对模型进行了训练和验证,并与基于历史平均值和卡尔曼滤波的行程时间预测模型进行对比.比较结果表明,LightGBM模型在MAE (Mean Absolute Error)和MAPE (Mean Absolute Percentage Error)这两个指标上均大幅度优于其他两个模型,说明LightGBM模型在公交车行程时间预测上具有很好的稳定性和应用前景.
In the operation of urban public transport networks, the travel time between stations of a bus is affected by road and environmental conditions. This paper analyzes the bus speed characteristics, road characteristics and weather characteristics during bus operation, and a feature-based LightGBM bus travel time prediction model is established. By adjusting the relevant parameters in the LightGBM algorithm, the weights of the influencing features and factors are assigned. Then the model is trained and verified by using the 24-day bus GPS data of one bus line in Tianjin, and compared with the travel time prediction model based on historical mean and Kalman filter. The comparison results show that the LightGBM model is superior to the other two models in MAE(Mean Absolute Error) and MAPE(Mean Absolute Percentage Error), indicating that the LightGBM model has good stability and application prospects in bus travel time prediction.
In the operation of urban public transport networks, the travel time between stations of a bus is affected by road and environmental conditions. This paper analyzes the bus speed characteristics, road characteristics and weather characteristics during bus operation, and a feature-based LightGBM bus travel time prediction model is established. By adjusting the relevant parameters in the LightGBM algorithm, the weights of the influencing features and factors are assigned. Then the model is trained and verified by using the 24-day bus GPS data of one bus line in Tianjin, and compared with the travel time prediction model based on historical mean and Kalman filter. The comparison results show that the LightGBM model is superior to the other two models in MAE(Mean Absolute Error) and MAPE(Mean Absolute Percentage Error), indicating that the LightGBM model has good stability and application prospects in bus travel time prediction.
引文
[1] RAMAKRISHNA Y, RAMAKRISHNA P,LAKSHMANAN V, et al. Use of GPS probe data and passenger data for prediction of bus transit travel time[C]//American Society of Civil Engineers.Transportation Land Use, Planning, and Air Quality Congress. Reston, VA:American Society of Civil Engineers, 2008:124-133.
[2]于滨,杨忠振,林剑艺.应用支持向量机预测公交车运行时间[J].系统工程理论与实践, 2007, 27(4):160-164,176.[YU B, YANG Z Z, LIN J Y. Bus arrival time prediction using support vector machines[J]. Systems Engineering-Theory&Practice, 2007, 27(4):160-164,176.]
[3] CHEN M, LIU X B, XIA J X, et al. A dynamic busarrival time prediction model based on APC data[J].Computer-Aided Civil and Infrastructure Engineering,2004, 19(5):364-376.
[4]陈旭梅,龚辉波,王景楠.基于SVM和Kalman滤波的BRT行程时间预测模型研究[J].交通运输系统工程与信息, 2012, 12(4):29-34.[CHEN X M, GONG H B,WANG J N. BRT vehicle travel time prediction based on SVM and Kalman filte[J]. Journal of Transportation Systems Engineering and Information Technology, 2012,12(4):29-34.]
[5] LIN Y J, YANG X F, ZOU N, et al. Real-time bus arrival time prediction:Case study for Jinan, China[J].Journal of Transportation Engineering, 2013, 139(11):1133-1140.
[6] DHIVYABHARATHI B, KUMAR B A, VANAJAKSHI L. Real time bus arrival time prediction system under Indian traffic condition[C]//2016 IEEE International Conference on Intelligent Transportation Engineering(ICITE 2016), New York, NY:IEEE Press, 2016:18-22.
[7] KE G L, MENG Q, FINLEY T, et al. LightGBM:A highly efficient gradient boosting decision tree[C]//Neural Information Processing Systems Foundation, Inc.Advances in Neural Information Processing Systems 30(NIPS 2017). San Diego, CA:Neural Information Processing Systems Foundation, Inc., 2017:Paper No.6907:1-9.
[8] WANG D H, ZHANG Y, ZHAO Y. LightGBM:An effective miRNA classification method in breast cancer patients[C]//Association for Computing Machinery.Proceedings of the 2017 International Conference on Computational Biology and Bioinformatics(ICCBB2017). New York, NY:Association for Computing Machinery, 2017:7-11.
[9]王殿海,汤月华,陈茜,等.基于GPS数据的公交站点区间行程时间可靠性影响因素[J].东南大学学报(自然科学版), 2015, 45(2):404-412.[WANG D H, TANG Y H,CHEN Q, et al. Influence factors of GPS-based bus travel time reliability between adjacent bus stations[J]. Journal of Southeast University(Natural Science Edition), 2015, 45(2):404-412.]
[10]寇俊佳.基于GPS的车辆到站实时预测模型的研究[D].南昌:南昌大学, 2015.[KOU J J. The research of bus real-time forecast arrival model based on GPS[D]. Nanchang:Nanchang University, 2015.]
[2]于滨,杨忠振,林剑艺.应用支持向量机预测公交车运行时间[J].系统工程理论与实践, 2007, 27(4):160-164,176.[YU B, YANG Z Z, LIN J Y. Bus arrival time prediction using support vector machines[J]. Systems Engineering-Theory&Practice, 2007, 27(4):160-164,176.]
[3] CHEN M, LIU X B, XIA J X, et al. A dynamic busarrival time prediction model based on APC data[J].Computer-Aided Civil and Infrastructure Engineering,2004, 19(5):364-376.
[4]陈旭梅,龚辉波,王景楠.基于SVM和Kalman滤波的BRT行程时间预测模型研究[J].交通运输系统工程与信息, 2012, 12(4):29-34.[CHEN X M, GONG H B,WANG J N. BRT vehicle travel time prediction based on SVM and Kalman filte[J]. Journal of Transportation Systems Engineering and Information Technology, 2012,12(4):29-34.]
[5] LIN Y J, YANG X F, ZOU N, et al. Real-time bus arrival time prediction:Case study for Jinan, China[J].Journal of Transportation Engineering, 2013, 139(11):1133-1140.
[6] DHIVYABHARATHI B, KUMAR B A, VANAJAKSHI L. Real time bus arrival time prediction system under Indian traffic condition[C]//2016 IEEE International Conference on Intelligent Transportation Engineering(ICITE 2016), New York, NY:IEEE Press, 2016:18-22.
[7] KE G L, MENG Q, FINLEY T, et al. LightGBM:A highly efficient gradient boosting decision tree[C]//Neural Information Processing Systems Foundation, Inc.Advances in Neural Information Processing Systems 30(NIPS 2017). San Diego, CA:Neural Information Processing Systems Foundation, Inc., 2017:Paper No.6907:1-9.
[8] WANG D H, ZHANG Y, ZHAO Y. LightGBM:An effective miRNA classification method in breast cancer patients[C]//Association for Computing Machinery.Proceedings of the 2017 International Conference on Computational Biology and Bioinformatics(ICCBB2017). New York, NY:Association for Computing Machinery, 2017:7-11.
[9]王殿海,汤月华,陈茜,等.基于GPS数据的公交站点区间行程时间可靠性影响因素[J].东南大学学报(自然科学版), 2015, 45(2):404-412.[WANG D H, TANG Y H,CHEN Q, et al. Influence factors of GPS-based bus travel time reliability between adjacent bus stations[J]. Journal of Southeast University(Natural Science Edition), 2015, 45(2):404-412.]
[10]寇俊佳.基于GPS的车辆到站实时预测模型的研究[D].南昌:南昌大学, 2015.[KOU J J. The research of bus real-time forecast arrival model based on GPS[D]. Nanchang:Nanchang University, 2015.]