基于IC卡信息的公交客流出行特征分析系统研究
|
摘要
为了最大限度发挥公共交通的优势,必须深入了解居民的出行需求和出行习惯,全面准确掌握公交客流。IC卡具有信息量丰富,数据全面可靠,技术成熟等的特点,在信息采集的过程中受人为因素影响极小,随着IC卡使用的普及以及数据挖掘技术的成熟,通过对IC卡数据库进行数据挖掘,可较容易得到可靠的公交客流信息。
虽然国内有不少通过IC卡获取客流出行特征的研究,但是由于对海量数据处理效率低下,以及IC卡和公交GPS系统结合不好,这些研究都只是提及理论与算法,对真正实现海量数据处理获取客流出行特征的研究仍然比较缺乏。正是在这样的背景下,本文探讨了实现基于IC卡信息获取公交客流出行特征的方法,并以系统的开发为最终成果,实现公交客流出行特征数据的统计分析处理。
本文首先介绍了IC卡数据的基本结构、公交GPS系统,以及IC卡信息与公交GPS系统的数据融合获取IC卡的客流信息以及线路各站点乘客上车人数的方法。然后详细介绍各种公交客流特征指标的统计模型,其中包括客流量统计指标统计模型、客流换乘统计指标统计模型、线路站点OD矩阵统计模型以及区域OD矩阵统计模型。本文创新点在于区域OD矩阵的统计中,采用IC卡匹配找出换乘乘客起讫点和线路OD,对基于出行理论统计方法推出的区域OD进行修正,该方法易于实现,能运用到实际中且运算效率高,适用于海量数据的客流出行特征统计分析。
在数据采集分析以及理论研究改进的基础上,完成基于IC卡数据的公交客流出行特征系统的开发,系统划分为客流相关指标统计模块、线路OD矩阵生成模块、区域OD矩阵生成模块三大模块。文章详细介绍各个模块的设计理念、设计流程以及模块功能。
最后,以广州市羊城通IC卡数据以及广州市公交系统信息为基础,应用开发的公交客流出行特征系统,统计广州市公交客流出行特征数据,同时分析系统运行情况,证明了系统具有运行效率高,实用性强,尤其适用于海量数据的客流出行特征统计的特点,根据所得结果详细分析目前广州市公交线网存在的问题,为今后调整优化线网提供了详实可靠的依据。
In order to maximize the advantages of public transport, we must better understand the travel demand and travel habits of residents, comprehensively and accurately master the volume of passengers. IC card has many features such as informative, comprehensive and technology maturity, minimal impact by human factors during the information gathering process, with the popularization of IC cards and data mining technology matures, it is easy to get more available and reliable public transport passenger information through the data mining of the IC card database.
Although there are many domestic research on passenger travel characteristics through the IC card, because of the low efficiency processing of the massive data, poor combined between the IC card and the public transport GPS system, these studies just mention the theory and algorithms, not really get the passenger travel characteristics from processing huge amounts of data. It is in this context, the paper discusses the method of getting bus passenger travel characteristics based on IC card, with the system development for the final results, achieve statistical analysis of bus passenger travel characteristics.
This paper introduces the basic structure of IC card data, bus GPS systems, data fusion between IC card information and public transportation GPS system to get passenger information and the number of site approach on bus lines. And then detail a variety of statistical model of bus passenger travel characteristics indicators, including the statistical models of passenger volume indicators, statistical models of passenger transfer indicators, line site OD Matrix statistical models, and regional OD Matrix statistical models. The innovation of the paper is the regional OD matrix statistics, using IC cards matches to find transfer passengers’OD, and amend the regional OD based on the travel theory. This method is easy to implement and high efficiency, can be applied to statistical analysis of passenger travel characteristics from huge amounts of data.
On the basis of data analysis and improving the theoretical research, complete the bus passenger travel characteristics system development based on IC card information. The system is divided into three modules, including passenger indicators statistics module, line site OD matrix generation module, regional OD matrix generation module. This paper details the design, process and the functions of each module.
Finally, based on the Guangzhou IC card data and information of Guangzhou public transport system, using the bus passenger travel characteristics system, add up the data of Guangzhou bus passenger travel characteristics, analyze the system operation to prove that the system has the features of efficient, practical, especially for passenger travel characteristics statistical from huge amounts of data. According to the results, analyze the current problems in public transportation network in Guangzhou, provides a reliable basis on adjusting and optimizing the transportation network for the future.
虽然国内有不少通过IC卡获取客流出行特征的研究,但是由于对海量数据处理效率低下,以及IC卡和公交GPS系统结合不好,这些研究都只是提及理论与算法,对真正实现海量数据处理获取客流出行特征的研究仍然比较缺乏。正是在这样的背景下,本文探讨了实现基于IC卡信息获取公交客流出行特征的方法,并以系统的开发为最终成果,实现公交客流出行特征数据的统计分析处理。
本文首先介绍了IC卡数据的基本结构、公交GPS系统,以及IC卡信息与公交GPS系统的数据融合获取IC卡的客流信息以及线路各站点乘客上车人数的方法。然后详细介绍各种公交客流特征指标的统计模型,其中包括客流量统计指标统计模型、客流换乘统计指标统计模型、线路站点OD矩阵统计模型以及区域OD矩阵统计模型。本文创新点在于区域OD矩阵的统计中,采用IC卡匹配找出换乘乘客起讫点和线路OD,对基于出行理论统计方法推出的区域OD进行修正,该方法易于实现,能运用到实际中且运算效率高,适用于海量数据的客流出行特征统计分析。
在数据采集分析以及理论研究改进的基础上,完成基于IC卡数据的公交客流出行特征系统的开发,系统划分为客流相关指标统计模块、线路OD矩阵生成模块、区域OD矩阵生成模块三大模块。文章详细介绍各个模块的设计理念、设计流程以及模块功能。
最后,以广州市羊城通IC卡数据以及广州市公交系统信息为基础,应用开发的公交客流出行特征系统,统计广州市公交客流出行特征数据,同时分析系统运行情况,证明了系统具有运行效率高,实用性强,尤其适用于海量数据的客流出行特征统计的特点,根据所得结果详细分析目前广州市公交线网存在的问题,为今后调整优化线网提供了详实可靠的依据。
In order to maximize the advantages of public transport, we must better understand the travel demand and travel habits of residents, comprehensively and accurately master the volume of passengers. IC card has many features such as informative, comprehensive and technology maturity, minimal impact by human factors during the information gathering process, with the popularization of IC cards and data mining technology matures, it is easy to get more available and reliable public transport passenger information through the data mining of the IC card database.
Although there are many domestic research on passenger travel characteristics through the IC card, because of the low efficiency processing of the massive data, poor combined between the IC card and the public transport GPS system, these studies just mention the theory and algorithms, not really get the passenger travel characteristics from processing huge amounts of data. It is in this context, the paper discusses the method of getting bus passenger travel characteristics based on IC card, with the system development for the final results, achieve statistical analysis of bus passenger travel characteristics.
This paper introduces the basic structure of IC card data, bus GPS systems, data fusion between IC card information and public transportation GPS system to get passenger information and the number of site approach on bus lines. And then detail a variety of statistical model of bus passenger travel characteristics indicators, including the statistical models of passenger volume indicators, statistical models of passenger transfer indicators, line site OD Matrix statistical models, and regional OD Matrix statistical models. The innovation of the paper is the regional OD matrix statistics, using IC cards matches to find transfer passengers’OD, and amend the regional OD based on the travel theory. This method is easy to implement and high efficiency, can be applied to statistical analysis of passenger travel characteristics from huge amounts of data.
On the basis of data analysis and improving the theoretical research, complete the bus passenger travel characteristics system development based on IC card information. The system is divided into three modules, including passenger indicators statistics module, line site OD matrix generation module, regional OD matrix generation module. This paper details the design, process and the functions of each module.
Finally, based on the Guangzhou IC card data and information of Guangzhou public transport system, using the bus passenger travel characteristics system, add up the data of Guangzhou bus passenger travel characteristics, analyze the system operation to prove that the system has the features of efficient, practical, especially for passenger travel characteristics statistical from huge amounts of data. According to the results, analyze the current problems in public transportation network in Guangzhou, provides a reliable basis on adjusting and optimizing the transportation network for the future.
引文
[1]陈鹏.车辆IC卡信息采集系统设计[D].北京:北京科技大学,2003.
[2] TANAKA MIKIO, SATO NORIO, SAKUMA YASUSHI, MAFUNE KAZUTOSHL. A Study on the Application of Data Mining Technology to the Analysis of Passenger Flow Data, Railway Technical Research Institute, 2002, 16, 11: 37-42.
[3] Avishai Ceder. Public transit planning and operation: theory, modelling and practice [M]. Elsevier 2007.
[4] LAWRENCE A.KLEIN. Sensor Technologies and Data Requirements for ITS [M]. Artech House,2001.
[5]Shawn M.Turner. Advanced Techniques for Travel Time Data Collection[J]. Transportation Research Board of the National Academies 2007.1.
[6]Alejandro angel, Mark hickman, Pitu Mirchandani, Dinesh chandnani. Methods of Traffic Data Collection using Aerial Video. The IEEE 5th International Conference on interlligent Transportation Systems, 2002.
[7]P.W.Bonsall. The changing face of parking-related data collection and analysis: The role of new technologies [J]. Springer Netherlands, 2004.1l.
[8]J.J.Barry, R Newhouser, A Rahbee, S.Sayeda. Origin and Destination Estimation in New York City with Automated Fare System Data [J]. Transportation Research Record, Vol.1817, 2002: 183-187.
[9]Jianhe Du, Lisa Aultman-Hall. Increasing the accuracy of trip rate information from passive multi-day GPS travel datasets: Automatic trip end identification issues[J]. Transportation Research Part A, 41, 2007: 220~232.
[10]于滨.基于IC卡收费系统的动态交通信息采集研究[D].吉林大学,2003
[11]戴霄.基于公交IC信息的公交数据分析方法研究[D].东南大学,2006
[12]郭捷.公交IC卡通勤乘客OD确定方法研究[D].东南大学,2006
[13]刘翠,陈洪仁.公交线路客流OD矩阵推算方法研究[J].城市交通,2007,5(4):81~84
[14]朱从坤,丁健霆,陈瑜.公交线路OD反推的结构化模型研究[J].哈尔滨工业大学学报,2005,37(6):851~853
[15]窦慧丽,刘好德,杨晓光.基于站点上下客人数的公交客流OD反推方法研究[J].交通与计算机,2007,25(2):79~82
[16]刘颖杰,靳文舟,康凯.基于IC信息和概率理论的公交OD反推方法[J].公路与汽运,2010,3:31~33
[17]李东明,胡鹏.公交线路客流OD反推研究[J].华中科技大学学报,2009,26(1):105~107
[18]陈素平,陈学武,杨敏.基于小票法的公交线路客流OD分析方法[J].交通信息与安全,2009,27(1):11~14
[19]朱晓宏,丁位东,孙泰屹.公交客流信息采集技术研究[J].运营与管理,2005,1:55~56
[20]舒国辉.基于数据挖掘的公交客流规律研究[D].北京交通大学,2009
[21]龙安国.基于GPS/GPRS的智能公交系统的设计与实现.通信技术. 2009, 42(1):326-330
[22]戴维.基于IC卡数据和公交车GPS信息的攻击OD矩阵推算[D].华中科技大学,2009
[23]陈学武,戴霄,陈茜.公交IC卡信息采集、分析与应用研究[J].土木工程学报,2004,37(2):105~110
[24]罗磊.基于IC卡信息的公交客流空间分布特征分析方法研究[D].东南大学,2009
[25]师富民.基于IC卡数据的公交OD矩阵构造方法研究[D].长春:吉林大学,2004.
[26]上海市城市交通规划研究所.客运规划与城市发展[R].上海:华东理工大学,1996.
[27]陈森发,周振国,于栋华.一种动态OD矩阵估算法的理论及应用[J]东南大学学报(自然科学版),2003,33(1):106~110;
[28]H.P.Lo,N.Zhang,W.H.K.Lam.Decomposition Algorithm for Statistical Estimation of OD Matrix with Random Link Choice Proportions form Traffic Counts[J].Transportation Research (PartB), 1999, (33): 369～385;
[29]S.C.Wong,C.O.Tong.Estimation of Time—Dependent Origin—Destination Matrices for Transit Networks[J].Transportation Research(PartB),1998,32(1):35～48
[30]尹娟,郭国会.人工神经网络在交通网络OD矩阵推算中的应用[J].浙江大学学报(工学版),2000,34(4):423～427
[31]俞洁,杨晓光.改进BP神经网络的公交线路OD矩阵推算方法[J].系统工程2006(4):89~92
[32]王炜,徐吉谦,杨涛,李旭宏.城市交通规划理论及其应用[M].南京:东南大学出版社,1998
[33]徐吉谦.交通工程总论[M].北京:人民交通出版社,2003.
[34]周涛,翟长旭,高志刚.基于公交IC卡数据的OD推算技术研究[J],城市交通,2007,5(3):48～52
[2] TANAKA MIKIO, SATO NORIO, SAKUMA YASUSHI, MAFUNE KAZUTOSHL. A Study on the Application of Data Mining Technology to the Analysis of Passenger Flow Data, Railway Technical Research Institute, 2002, 16, 11: 37-42.
[3] Avishai Ceder. Public transit planning and operation: theory, modelling and practice [M]. Elsevier 2007.
[4] LAWRENCE A.KLEIN. Sensor Technologies and Data Requirements for ITS [M]. Artech House,2001.
[5]Shawn M.Turner. Advanced Techniques for Travel Time Data Collection[J]. Transportation Research Board of the National Academies 2007.1.
[6]Alejandro angel, Mark hickman, Pitu Mirchandani, Dinesh chandnani. Methods of Traffic Data Collection using Aerial Video. The IEEE 5th International Conference on interlligent Transportation Systems, 2002.
[7]P.W.Bonsall. The changing face of parking-related data collection and analysis: The role of new technologies [J]. Springer Netherlands, 2004.1l.
[8]J.J.Barry, R Newhouser, A Rahbee, S.Sayeda. Origin and Destination Estimation in New York City with Automated Fare System Data [J]. Transportation Research Record, Vol.1817, 2002: 183-187.
[9]Jianhe Du, Lisa Aultman-Hall. Increasing the accuracy of trip rate information from passive multi-day GPS travel datasets: Automatic trip end identification issues[J]. Transportation Research Part A, 41, 2007: 220~232.
[10]于滨.基于IC卡收费系统的动态交通信息采集研究[D].吉林大学,2003
[11]戴霄.基于公交IC信息的公交数据分析方法研究[D].东南大学,2006
[12]郭捷.公交IC卡通勤乘客OD确定方法研究[D].东南大学,2006
[13]刘翠,陈洪仁.公交线路客流OD矩阵推算方法研究[J].城市交通,2007,5(4):81~84
[14]朱从坤,丁健霆,陈瑜.公交线路OD反推的结构化模型研究[J].哈尔滨工业大学学报,2005,37(6):851~853
[15]窦慧丽,刘好德,杨晓光.基于站点上下客人数的公交客流OD反推方法研究[J].交通与计算机,2007,25(2):79~82
[16]刘颖杰,靳文舟,康凯.基于IC信息和概率理论的公交OD反推方法[J].公路与汽运,2010,3:31~33
[17]李东明,胡鹏.公交线路客流OD反推研究[J].华中科技大学学报,2009,26(1):105~107
[18]陈素平,陈学武,杨敏.基于小票法的公交线路客流OD分析方法[J].交通信息与安全,2009,27(1):11~14
[19]朱晓宏,丁位东,孙泰屹.公交客流信息采集技术研究[J].运营与管理,2005,1:55~56
[20]舒国辉.基于数据挖掘的公交客流规律研究[D].北京交通大学,2009
[21]龙安国.基于GPS/GPRS的智能公交系统的设计与实现.通信技术. 2009, 42(1):326-330
[22]戴维.基于IC卡数据和公交车GPS信息的攻击OD矩阵推算[D].华中科技大学,2009
[23]陈学武,戴霄,陈茜.公交IC卡信息采集、分析与应用研究[J].土木工程学报,2004,37(2):105~110
[24]罗磊.基于IC卡信息的公交客流空间分布特征分析方法研究[D].东南大学,2009
[25]师富民.基于IC卡数据的公交OD矩阵构造方法研究[D].长春:吉林大学,2004.
[26]上海市城市交通规划研究所.客运规划与城市发展[R].上海:华东理工大学,1996.
[27]陈森发,周振国,于栋华.一种动态OD矩阵估算法的理论及应用[J]东南大学学报(自然科学版),2003,33(1):106~110;
[28]H.P.Lo,N.Zhang,W.H.K.Lam.Decomposition Algorithm for Statistical Estimation of OD Matrix with Random Link Choice Proportions form Traffic Counts[J].Transportation Research (PartB), 1999, (33): 369～385;
[29]S.C.Wong,C.O.Tong.Estimation of Time—Dependent Origin—Destination Matrices for Transit Networks[J].Transportation Research(PartB),1998,32(1):35～48
[30]尹娟,郭国会.人工神经网络在交通网络OD矩阵推算中的应用[J].浙江大学学报(工学版),2000,34(4):423～427
[31]俞洁,杨晓光.改进BP神经网络的公交线路OD矩阵推算方法[J].系统工程2006(4):89~92
[32]王炜,徐吉谦,杨涛,李旭宏.城市交通规划理论及其应用[M].南京:东南大学出版社,1998
[33]徐吉谦.交通工程总论[M].北京:人民交通出版社,2003.
[34]周涛,翟长旭,高志刚.基于公交IC卡数据的OD推算技术研究[J],城市交通,2007,5(3):48～52