公交车辆到站时间预测研究
|
摘要
公交车辆到站时间是出行者最为关心的交通信息之一,提高公交车辆到站时间预测模型的精度和可靠性,可对城市公共交通的发展起到积极的推动作用。
论文系统地分析了公交车辆到站时间的组成部分和影响因素,选取公交车辆在前续站点的到站时间、停靠时间和延误程度为预测模型的输入变量,设计并实现了基于车辆GPS数据的插值方法、数据库结构和数据处理算法,得到了公交车辆在每站的详细运行数据。
在此基础之上,论文首先提出了基于平均行驶和停靠时间的统计模型；随后建立了BP人工神经网络预测模型,使用样本数据来训练神经网络,拟合前续站点到站时间、停靠时间和延误程度与后续到站时间之间的非线性关系,利用实时的已到站信息对后续到站时间进行预测；最后提出了改进的非参数回归模型,先对搜索数据库进行聚类分析,随后对状态向量进行主成分分析以达到降维的效果,在此基础之上选取K近邻机制和加权平均预测算法来构建模型。
最后论文对北京公交16路实际采集来的3369组数据进行实证分析,以上行方向为例,详细分析了数据特性,选取相对平均误差MRE作为评价模型预测效果的指标,对三个模型进行了计算,发现非参数回归预测模型在最优K值条件下的平均预测误差最小,相对于统计模型和人工神经网络模型分别改善了55.15%和40.24%。
Bus arrival time is one of the most concerned traffic information for travelers and it is a very important subject to improve the precision and reliability of the prediction model which can promote the development of city public transportation.
This thesis analyzes the components of bus arrival time systematically and the bus arrival time, dwell time and stop delay at previous stops are chosen as the main input variables of the prediction model. After that this thesis designs the algorithm of data interpolation and processing in order to get the input variables of the prediction models.
Based on the processed data and analysis, this thesis firstly puts forwards the statistical model based on average running time of each link and dwelling time of each stop. After that BP-Artificial Neural Network (BP-ANN) is modeled and the mass sample data are used to train the non-linear relationships between the input and output variables. With the trained network and real-time input variables of the previous stops, the bus arrival time of the coming stops can be forecasted. In the end, this thesis uses the improved non-parameter regression method by using the principal component analysis and clustering methods. Then the specified procedures and steps of the improved model are designed by using K nearest neighbor mechanism and weighted arithmetic mean method.
In the end this thesis gives the case study of the 3369 group data from line 16 of Beijing Bus Company. It analyzes the data characteristics and uses the index of Mean Relative Error (MRE) to evaluate the three models. It is founded that the non-parameter regression model of the optimum K value can obtain the best average prediction result which has improved by 55.15% and 40.24% compared to statistical model and BP-ANN respectively.
论文系统地分析了公交车辆到站时间的组成部分和影响因素,选取公交车辆在前续站点的到站时间、停靠时间和延误程度为预测模型的输入变量,设计并实现了基于车辆GPS数据的插值方法、数据库结构和数据处理算法,得到了公交车辆在每站的详细运行数据。
在此基础之上,论文首先提出了基于平均行驶和停靠时间的统计模型；随后建立了BP人工神经网络预测模型,使用样本数据来训练神经网络,拟合前续站点到站时间、停靠时间和延误程度与后续到站时间之间的非线性关系,利用实时的已到站信息对后续到站时间进行预测；最后提出了改进的非参数回归模型,先对搜索数据库进行聚类分析,随后对状态向量进行主成分分析以达到降维的效果,在此基础之上选取K近邻机制和加权平均预测算法来构建模型。
最后论文对北京公交16路实际采集来的3369组数据进行实证分析,以上行方向为例,详细分析了数据特性,选取相对平均误差MRE作为评价模型预测效果的指标,对三个模型进行了计算,发现非参数回归预测模型在最优K值条件下的平均预测误差最小,相对于统计模型和人工神经网络模型分别改善了55.15%和40.24%。
Bus arrival time is one of the most concerned traffic information for travelers and it is a very important subject to improve the precision and reliability of the prediction model which can promote the development of city public transportation.
This thesis analyzes the components of bus arrival time systematically and the bus arrival time, dwell time and stop delay at previous stops are chosen as the main input variables of the prediction model. After that this thesis designs the algorithm of data interpolation and processing in order to get the input variables of the prediction models.
Based on the processed data and analysis, this thesis firstly puts forwards the statistical model based on average running time of each link and dwelling time of each stop. After that BP-Artificial Neural Network (BP-ANN) is modeled and the mass sample data are used to train the non-linear relationships between the input and output variables. With the trained network and real-time input variables of the previous stops, the bus arrival time of the coming stops can be forecasted. In the end, this thesis uses the improved non-parameter regression method by using the principal component analysis and clustering methods. Then the specified procedures and steps of the improved model are designed by using K nearest neighbor mechanism and weighted arithmetic mean method.
In the end this thesis gives the case study of the 3369 group data from line 16 of Beijing Bus Company. It analyzes the data characteristics and uses the index of Mean Relative Error (MRE) to evaluate the three models. It is founded that the non-parameter regression model of the optimum K value can obtain the best average prediction result which has improved by 55.15% and 40.24% compared to statistical model and BP-ANN respectively.
引文
[1]陈石.区域公交时刻表生成的模型与方法研究[D].北京交通大学硕士学位论文,2009.
[2]罗虹.基于GPS的公交车辆到达时间预测技术研究[D].重庆大学硕士学位论文,2007.
[3]彭聪.基于GIS的城市公共交通监控调度系统数据库的研究[D].中国科学院广州地球化学研究所,2005.
[4]张斐斐.公共交通驾驶员调度问题研究[D].北京交通大学硕士学位论文,2006.
[5]张飞舟,杨东凯,范跃祖，孙先仿.智能交通系统中的公共交通信息管理系统[J].北京航空航天大学学报,2000,26(4)：385-388.
[6]Bae,S.,and P.Kachroo. Proactive Travel Time Predictions Under Interrupted Flow Condition. CD-ROM. In Proceedings of the 6th International Vehicle Navigation & Information Systems, Seatle, Washington,1995.
[7]Chen M.,Chien S.I.,Liu X.,Brickey J.M.Application of APC/AVL Archived Data Support System. In TRB 82nd Annual Meeting (CD-ROM),Washington,D.C.2003.
[8]Lin, W., and J. Zeng. An Experimental Study on Real Time Bus Arrival Time Prediction with GPS Data. CD-ROM.78th Annual Meeting of the Transportation Research Board, National Research Council, Washington D.C., January 1999.
[9]Patnaik J.,Chien S.,and Bladikas A. Estimation of Bus Arrival Times Using APC Data. Journal of Public Transportation,2004,7(1):1-20.
[10]周雪梅,杨晓光,王磊.公交车辆行程时间预测方法研究[J].交通与计算机,2002,20(6)：12-14.
[11]杨兆升,保丽霞,朱国华.基于Fuzzy回归的快速路行程时间预测模型研究[J].公路交通科技,2004,21(3)：78-80.
[12]D.Angelo.M.P.,H.M.Al-Deek and M.C.Wang. Travel Time Prediction for Freeway Corridors. In TRB 78th Annual Meeting (CD-ROM), Washington D.C.,1999.
[13]Chien, Steven I.J., Ding Y., and Wei C.Dynamic Bus Arrival Time Prediction with Artificial Neural Networks.Journal of Transportation Engineering,2002,128(5):429-438.
[14]于滨,杨中振,曾庆成.基于SVM和Kamlan滤波的公交车到站时间预测模型[J].2008,21(2)：89-92.
[15]Wall, Z., and D. J. Dailey. Algorithm for Predicting the Arrival Time of Mass Transit Vehicles Using Automatic Vehicle Location Data. In 78th Annual Meeting of the Transportation Research Board, Washington, D.C.,1999.
[16]CATHEY F W, DAILIEY D J. A Prescription for Transit Arrival/Departure Prediction Using Automatic Vehicle Location Data[J]. Transportation Research Part C,2003,11(3): 241-264.
[17]Shalaby, A. S., and A. Farhan. Bus Travel Time Prediction for Dynamic Operations Control and Passenger Information Systems. In 82ndAnnual Meeting of the Transportation Research Board, Washington, D.C.,2003.
[18]温惠英,徐建闽,傅惠.基于灰色关联分析的路段行程时间卡尔曼滤波预测算法[J].华南理工大学学报(自然科学版),2006,34(9)：66-69.
[19]Jeong R.and Rilett L.R.Bus Arrival Time Prediction Using Artificial Neural Network Model.In IEEE 7th Intelligent Transportation Systems Conference, Washington D.C.,2004, 988-993.
[20]张堂贤,郭中天.公交车到站时间暨复合路线旅行时间预估模式的研究[J].土木工程学报,2005,38(12)：115-123.
[21]杨兆升.城市智能公共交通系统理论与方法[M].北京：中国铁道出版社,2004.
[22]何启海,方钰.基于PDA的上海市交通信息网格发布平台[J].计算机工程.2006,32(1):242-244.
[23]周小蓉,鲍轶洲,李红宝.广‘州市先进的公共交通系统工程建设实践[J].城市交通,2009,5：86-90.
[24]Avishai Ceder(以)著,关伟等译.公共交通规划与运营：理论、建模及应用.北京：清华大学出版社,2010.
[25]邓晖,吴笃斌.GPS,GIS,GSM技术在特种服务车辆监控中的应用[J].电子技术.2001(7):22-24.
[26]高博.车辆导航系统中数据处理、地图匹配和路径规划的研究[D].解放军信息工程大学硕士学位论文,2001.
[27]侯晋瑞.车辆监控系统监控中心端设计与实现[D].西南交通大学硕士学位论文,2007.
[28]王俊,陈学武.影响城市公交车辆运行时间的因素分析及改进措施[J].城市公共交通,2004,1(1)：6-7.
[29]百度百科.ittp://baike.baidu.com/
[30]D.Park and L.R.Rilett.Forecasting Multi-Period Freeway Link Travel Times Using Modular Neural Networks. In Transportation Research Record:Journal of Transportation Research Board, No.1617, National Research Council, Washington, D.C.,1998,163-170.
[31]张立明.人工神经网络的模型及其应用[M].上海：复旦大学出版社,1994.
[32]宫晓燕,汤淑明.基于非参数回归的短时交通流预测与事件检测综合算法[J].中国公路学报，2003,16(1)：82-86.
[33]Smith B L, Demetsky M J. Short-term traffic flow prediction models-A comparison of neural network and nonparametric regression approaches [A]. In:Proceedings of IEEE International Conference on Systems[C],San Antonio,TX,USA:IEEE.1994:1706-1709.
[34]贺国光.ITS系统工程导论[M].北京：中国铁道出版社,2004.
[35]张晓利,贺国光,陆化普.基于K-邻域非参数回归短时交通流预测方法[J].系统工程学报,2009,24(2)：178-183.
[36]范鲁明.基于非参数回归的短时交通流量预测[D].天津大学硕士学位论文,2007.
[37]Ran Hee Jeong.The Prediction of Bus Arrival Time Using Automatic Vehicle Location System Data [Dissertation], Texas A&M University,2004.
[38]陈鹏.基于BP神经网络的公交智能实时调度模型研究及系统实现[D].北京交通大学硕士学位论文,2008.
[39]翁剑成,荣建,任福田,魏中华.基于非参数回归的快速路行程速度短期预测算法[J],公路交通科技,2007,24(3)：93-97.
[40]张晓利.基于非参数回归的短时交通流量预测方法研究[D],天津大学博士学位论文,2007.
[41]米红,张文璋.实用现代统计分析方法与SPSS应用[M].北京：当代中国出版社,2000.
[42]Richard A.Johnson, Dean W.Wichern.Applied Multivariate Statistical Analysis [M].Person Prentice Hall 2001.4:347-348.
[43]荣秋生,颜君彪,郭国强.基于DBSCAN聚类算法的研究与实现[J].计算机应用,2004,1(24)：45-46.
[44]田大东,邓伟.改进的K均值聚类算法在支持矢量机中的应用[J].计算机工程与应用,2007,43(32)：161-163.
[45]周水庚,周傲英,曹晶,胡运发.一种基于密度的快速聚类算法[J].计算机研究与发展,2000,37(11)：1287-1292.
[46]徐义田,王来生,张好治,孙宝山.基于SVM的分类算法与聚类分析[J].烟台大学学报(自然科学与工程版),2004,17(1)：10-13.
[47]Xia Shixiong,Li Wenchao,Zhou Yong,Zhang Lei,Niu Qiang.Improved K-means Clustering Algorithm[J],Journal of Southeast University(English Edition) 2007,23(3):435-438.
[48]朱娴,马卫.一种基于层次聚类的双聚类算法[J],微计算机应用,2009,30(5)：12-16.
[49]Jiawei Han,Micheline Kamber(加)著,范明,孟小峰译.数据挖掘概念与技术[M].北京：机械工业出版社,2008.
[50]张涛,陈先,谢美萍,张玥杰.基于K近邻非参数回归的短时交通流预测方法[J].系统工程理论与实践,2010,30(2)：376-384.
[51]MapInfo公司.MapXtreme 2005开发人员指南[M].2007.
[52]薛华成.管理信息系统[M].北京：清华大学出版社,2005.
[53]胡振文，孙玉梅,李仁杰.地理信息系统原理与应用[M].北京：中国铁道出版社,2005.
[54]Karli Watson,Christian Nagel(美)等著,齐立波,黄静译.C#入门经典.北京：清华大学出
版社，2005.
[55]牛志广,张宏伟,辛志伟.基于log-logistic概率分布的近海水质组合预测方法研究[J].
系统工程理论与实践,2006,5(5)：111-116.
[56]陈杰.Matlab宝典[M].北京：电子工业出版社,2007.
[57]刘智勇.SQL Server 2005宝典[M].北京：电子工业出版社,2007.
[2]罗虹.基于GPS的公交车辆到达时间预测技术研究[D].重庆大学硕士学位论文,2007.
[3]彭聪.基于GIS的城市公共交通监控调度系统数据库的研究[D].中国科学院广州地球化学研究所,2005.
[4]张斐斐.公共交通驾驶员调度问题研究[D].北京交通大学硕士学位论文,2006.
[5]张飞舟,杨东凯,范跃祖，孙先仿.智能交通系统中的公共交通信息管理系统[J].北京航空航天大学学报,2000,26(4)：385-388.
[6]Bae,S.,and P.Kachroo. Proactive Travel Time Predictions Under Interrupted Flow Condition. CD-ROM. In Proceedings of the 6th International Vehicle Navigation & Information Systems, Seatle, Washington,1995.
[7]Chen M.,Chien S.I.,Liu X.,Brickey J.M.Application of APC/AVL Archived Data Support System. In TRB 82nd Annual Meeting (CD-ROM),Washington,D.C.2003.
[8]Lin, W., and J. Zeng. An Experimental Study on Real Time Bus Arrival Time Prediction with GPS Data. CD-ROM.78th Annual Meeting of the Transportation Research Board, National Research Council, Washington D.C., January 1999.
[9]Patnaik J.,Chien S.,and Bladikas A. Estimation of Bus Arrival Times Using APC Data. Journal of Public Transportation,2004,7(1):1-20.
[10]周雪梅,杨晓光,王磊.公交车辆行程时间预测方法研究[J].交通与计算机,2002,20(6)：12-14.
[11]杨兆升,保丽霞,朱国华.基于Fuzzy回归的快速路行程时间预测模型研究[J].公路交通科技,2004,21(3)：78-80.
[12]D.Angelo.M.P.,H.M.Al-Deek and M.C.Wang. Travel Time Prediction for Freeway Corridors. In TRB 78th Annual Meeting (CD-ROM), Washington D.C.,1999.
[13]Chien, Steven I.J., Ding Y., and Wei C.Dynamic Bus Arrival Time Prediction with Artificial Neural Networks.Journal of Transportation Engineering,2002,128(5):429-438.
[14]于滨,杨中振,曾庆成.基于SVM和Kamlan滤波的公交车到站时间预测模型[J].2008,21(2)：89-92.
[15]Wall, Z., and D. J. Dailey. Algorithm for Predicting the Arrival Time of Mass Transit Vehicles Using Automatic Vehicle Location Data. In 78th Annual Meeting of the Transportation Research Board, Washington, D.C.,1999.
[16]CATHEY F W, DAILIEY D J. A Prescription for Transit Arrival/Departure Prediction Using Automatic Vehicle Location Data[J]. Transportation Research Part C,2003,11(3): 241-264.
[17]Shalaby, A. S., and A. Farhan. Bus Travel Time Prediction for Dynamic Operations Control and Passenger Information Systems. In 82ndAnnual Meeting of the Transportation Research Board, Washington, D.C.,2003.
[18]温惠英,徐建闽,傅惠.基于灰色关联分析的路段行程时间卡尔曼滤波预测算法[J].华南理工大学学报(自然科学版),2006,34(9)：66-69.
[19]Jeong R.and Rilett L.R.Bus Arrival Time Prediction Using Artificial Neural Network Model.In IEEE 7th Intelligent Transportation Systems Conference, Washington D.C.,2004, 988-993.
[20]张堂贤,郭中天.公交车到站时间暨复合路线旅行时间预估模式的研究[J].土木工程学报,2005,38(12)：115-123.
[21]杨兆升.城市智能公共交通系统理论与方法[M].北京：中国铁道出版社,2004.
[22]何启海,方钰.基于PDA的上海市交通信息网格发布平台[J].计算机工程.2006,32(1):242-244.
[23]周小蓉,鲍轶洲,李红宝.广‘州市先进的公共交通系统工程建设实践[J].城市交通,2009,5：86-90.
[24]Avishai Ceder(以)著,关伟等译.公共交通规划与运营：理论、建模及应用.北京：清华大学出版社,2010.
[25]邓晖,吴笃斌.GPS,GIS,GSM技术在特种服务车辆监控中的应用[J].电子技术.2001(7):22-24.
[26]高博.车辆导航系统中数据处理、地图匹配和路径规划的研究[D].解放军信息工程大学硕士学位论文,2001.
[27]侯晋瑞.车辆监控系统监控中心端设计与实现[D].西南交通大学硕士学位论文,2007.
[28]王俊,陈学武.影响城市公交车辆运行时间的因素分析及改进措施[J].城市公共交通,2004,1(1)：6-7.
[29]百度百科.ittp://baike.baidu.com/
[30]D.Park and L.R.Rilett.Forecasting Multi-Period Freeway Link Travel Times Using Modular Neural Networks. In Transportation Research Record:Journal of Transportation Research Board, No.1617, National Research Council, Washington, D.C.,1998,163-170.
[31]张立明.人工神经网络的模型及其应用[M].上海：复旦大学出版社,1994.
[32]宫晓燕,汤淑明.基于非参数回归的短时交通流预测与事件检测综合算法[J].中国公路学报，2003,16(1)：82-86.
[33]Smith B L, Demetsky M J. Short-term traffic flow prediction models-A comparison of neural network and nonparametric regression approaches [A]. In:Proceedings of IEEE International Conference on Systems[C],San Antonio,TX,USA:IEEE.1994:1706-1709.
[34]贺国光.ITS系统工程导论[M].北京：中国铁道出版社,2004.
[35]张晓利,贺国光,陆化普.基于K-邻域非参数回归短时交通流预测方法[J].系统工程学报,2009,24(2)：178-183.
[36]范鲁明.基于非参数回归的短时交通流量预测[D].天津大学硕士学位论文,2007.
[37]Ran Hee Jeong.The Prediction of Bus Arrival Time Using Automatic Vehicle Location System Data [Dissertation], Texas A&M University,2004.
[38]陈鹏.基于BP神经网络的公交智能实时调度模型研究及系统实现[D].北京交通大学硕士学位论文,2008.
[39]翁剑成,荣建,任福田,魏中华.基于非参数回归的快速路行程速度短期预测算法[J],公路交通科技,2007,24(3)：93-97.
[40]张晓利.基于非参数回归的短时交通流量预测方法研究[D],天津大学博士学位论文,2007.
[41]米红,张文璋.实用现代统计分析方法与SPSS应用[M].北京：当代中国出版社,2000.
[42]Richard A.Johnson, Dean W.Wichern.Applied Multivariate Statistical Analysis [M].Person Prentice Hall 2001.4:347-348.
[43]荣秋生,颜君彪,郭国强.基于DBSCAN聚类算法的研究与实现[J].计算机应用,2004,1(24)：45-46.
[44]田大东,邓伟.改进的K均值聚类算法在支持矢量机中的应用[J].计算机工程与应用,2007,43(32)：161-163.
[45]周水庚,周傲英,曹晶,胡运发.一种基于密度的快速聚类算法[J].计算机研究与发展,2000,37(11)：1287-1292.
[46]徐义田,王来生,张好治,孙宝山.基于SVM的分类算法与聚类分析[J].烟台大学学报(自然科学与工程版),2004,17(1)：10-13.
[47]Xia Shixiong,Li Wenchao,Zhou Yong,Zhang Lei,Niu Qiang.Improved K-means Clustering Algorithm[J],Journal of Southeast University(English Edition) 2007,23(3):435-438.
[48]朱娴,马卫.一种基于层次聚类的双聚类算法[J],微计算机应用,2009,30(5)：12-16.
[49]Jiawei Han,Micheline Kamber(加)著,范明,孟小峰译.数据挖掘概念与技术[M].北京：机械工业出版社,2008.
[50]张涛,陈先,谢美萍,张玥杰.基于K近邻非参数回归的短时交通流预测方法[J].系统工程理论与实践,2010,30(2)：376-384.
[51]MapInfo公司.MapXtreme 2005开发人员指南[M].2007.
[52]薛华成.管理信息系统[M].北京：清华大学出版社,2005.
[53]胡振文，孙玉梅,李仁杰.地理信息系统原理与应用[M].北京：中国铁道出版社,2005.
[54]Karli Watson,Christian Nagel(美)等著,齐立波,黄静译.C#入门经典.北京：清华大学出
版社，2005.
[55]牛志广,张宏伟,辛志伟.基于log-logistic概率分布的近海水质组合预测方法研究[J].
系统工程理论与实践,2006,5(5)：111-116.
[56]陈杰.Matlab宝典[M].北京：电子工业出版社,2007.
[57]刘智勇.SQL Server 2005宝典[M].北京：电子工业出版社,2007.