城市轨道交通信息融合与决策方法研究
|
摘要
随着城市社会经济的发展和城市化进程的加快，城市人口急剧增加，大量流动人口进入城市，人员出行和物资交流频繁，城市交通形势严峻。世界各国的大中城市普遍存在着道路拥堵、交通秩序混乱的现象，城市交通矛盾越来越突出，成为制约城市经济发展的主要障碍。为解决城市交通拥堵问题，国内外很多大城市相继提出了以轨道交通为骨干、公共交通为主体的城市客运交通发展模式，但资金、政策、环境、决策理念等诸多因素却制约着我国城市轨道交通的良好发展。因此，结合我国城市特征和城市轨道交通运营特点，对城市轨道交通系统选型、运营调度方案和政策性亏损补贴等问题进行研究对我国城市轨道交通决策具有重要的现实意义和理论价值。
本文结合我国城市轨道交通的发展现状，分析城市轨道交通系统内各子系统的信息化程度以及信息的完整性，对城市轨道交通系统选型、运营调度方案和政策性亏损补贴等问题的理论和方法开展了探索性研究。研究期望得出一套科学的、系统化的城市轨道交通运营决策的理论和方法体系，确定城市轨道交通系统的补贴方式，明确财政补贴的额度；完善数据采集方法，进一步实行数据挖掘，提高数据的利用率，实现动态实时调度，提高城市轨道交通运输效率；明确城市轨道交通中各条线路的功能定位和性能标准，并选择与之相适应的系统型式，以满足城市发展和城市交通的要求，辅助各城市进行城市轨道交通的规划决策，并为国家管理决策部门的决策管理提供依据，保障我国城市轨道交通的可持续发展。研究中运用信息融合理论和技术方法对城市轨道交通系统选型、运营调度方案和政策性亏损补贴等问题进行了理论上的探索和决策方法的研究。具体研究内容和创新性成果如下：
(1)在综合分析信息融合理论和城市轨道交通信息化内容的基础上，提出了城市轨道交通信息融合理论体系，阐述了该理论体系的概念定义、核心理念、从属关系等内容；将城市轨道交通决策问题分为数据层融合、特征层融合和决策层融合三个层次，明确了各个层次决策的输入输出单元类型、决策目标以及需要研究解决的内容和相互间的逻辑从属关系，并针对各个层次给出了对应的实际决策问题，分别为基于数据融合的城市轨道交通财政补贴决策，基于特征融合的城市轨道交通实时调度决策以及基于决策融合的城市轨道交通系统型式选择决策。
(2)对城市轨道主要的技术指标进行了汇总介绍。将城市轨道交通财政补贴问题分解为财政补贴方式选择和财政补贴决策两个阶段。基于信息融合的理论和方法，将城市轨道交通各个系统直接采集到的数据进行相关和集成融合计算出客流量、平均发车间隔和平均运距等。以社会福利最大化为目标，以票价、发车间隔、站间距以及服务区域面积等为决策变量，对城市轨道交通财政补贴问题进行了优化。计算结果表明，在社会福利最大化的目标下，收支平衡财政补贴政策适合于城市轨道交通系统。在此基础上，通过对车辆运营成本与乘客等车时间成本的分析，确定了乘客总的等车时间成本是常量固定的社会成本。利用边际成本法建立了城市轨道交通补贴决策模型。最后结合长春市轻轨的运营数据进行了实证分析，经检验证明该模型对城市轨道交通财政补贴具有辅助决策作用。
(3)从城市轨道交通实时调度决策问题的分析入手，探寻调度决策问题的实质，进而总结出现有数据、方法和技术的不足。为完善原始数据的完整性，提出了乘客出行讫点明确性和TOD。探讨了新的数据采集方法，给出了对现阶段售票系统的改进方案，从而可以实时的采集到更完整的乘客出行信息，使实时调度的实现成为可能。基于信息融合理论的基本思想，将历史数据中的OD与车辆运行时间线相融合，得出TOD进而计算出客流数据的特征系数。并且以后车与前车的发车间隔作为模型的变量，以发车间隔最大为目标，同时考虑最小发车间隔、乘客所能容忍的最大等车时间和车厢容量的约束建立模型，得出单方向实时的非均匀的发车时刻表，作为车辆调度的依据。
(4)将证据理论引入城市轨道交通系统型式选择决策问题，把城市轨道交通系统型式选择问题定义为：以应用各个城市指标确定的各种系统型式的概率赋值作为证据，再将这些证据按照一定的组合规则进行组合做出城市轨道交通系统型式选择决策的融合问题。根据D-S证据理论的推理方法，将城市轨道交通系统型式选择决策定义为论域；城市轨道交通系统的型式定义为论域内的元素；与城市轨道交通系统建设决策有关的指标定义为证据；将各条证据进行组合得出城市轨道交通系统型式选择的综合概率支持函数，从而为城市轨道交通系统型式选择做出决策。据此，建立了城市轨道交通系统型式选择模型，探讨了适用于城市轨道交通系统型式选择问题的组合规则，并应用所得证据组合方法进行求解。
本文的研究成果将丰富与拓展城市轨道交通运营与发展决策方法的理论框架与方法体系，为城市轨道交通运营与发展决策提供理论依据。
With the acceleration of the urban social and economic development and urbanizationprocess, there is a sharp increase in urban population, a large number of population floatinto the city, Staff travel and commodity exchange occur frequently,so the situation of theurban traffic is becoming grim. Many large and medium cities of the world have theproblem of road congestion and traffic disorder, the contradiction of the urban traffic ismore and more prominent, which lead to restrict the development of urban economy. Manybig cities at home and abroad have pointed out the development model of urban passengertransport which based on the urban rail transit and public transport in order to address theproblem of urban traffic congestion. But many factors, such as funding, environment,decision-making and policy, are restricting the development of urban rail transit.So it hasimportant practical and theoretical value to study on the urban rail transit system selection,operation scheduling scheme and policy subsidies for losses.
Based on the current development of China’s urban rail transport, this paper analysedthe degree and the integrity of the information of each subsystem in the urban rail transitsystem, and carried out exploratory research on the theories and methods of urban railtransit system selection, operation scheduling scheme and Policy subsidies for losses. A setof theory and methodology of operational decision-making for urban rail transit which isscientific and systematic was expected to draw to ascertain the subsidies for urban railtransit system and the amount of financial subsidies, to improve the data collectionmethods which can do further data mining and improve the utilization of data, to achievedynamic real-time scheduling in order to improve the efficiency of urban rail transit, toclear the functional position and performance standards of urban rail transit lines and selectthe corresponding system type in order to meet the requirements of urban transportationand make planning decisions of urban rail transit in cities,and to provide the basis fordecision management of the state department. The paper made theoretical exploration anddecision-making method on urban rail transit system selection, Operation schedulingscheme and Policy subsidies for losses by using the theory and techniques of informationfusion. Specific research and innovation results are given in the following:
(1) Based on the the theory of information fusion and the information of urban rail transit, This paper pointed out the theoretical system for information fusion of urban railtransit and elaborated the conception,core concept and affiliation of the theoretical system;The information fusion of urban rail transit was divided into three levels: datafusion,feature fusion and decision fusion,then cleared the type of input and output unit foreach level decision-making,and each level has correspnding decision-making: financialsubsidy decisions of urban rail transit based on data fusion, real-time scheduling decisionsof urban rail transport based on feature fusion and type selection decisions of urban railtransit system based on decision fusion.
(2) The paper made summary introduction about the main technical indicators ofurban rail transit.data from each system of urban rail transit was collected and fused tocalculate the passenger traffic, the average grid spacing and average distance. Aimed atmaximizing the social welfare, the paper optimized the financial subsidies of urban railtransit according to the decision variables such as the fare, the grid spacing, station spacingand service area, etc. The calculation results showed that break-even financial subsidypolicy was suitable for urban rail transit system to the goal of maximizing social welfare.On this basis, it could be identified that the total waiting time cost of passengers wasconstant fixed social cost by analyzing vehicle operating cost and waiting time cost ofpassengers, and made the decision model of urban rail transit subsidy decision by usingmarginal cost method.Finally analysized the operational data of light rail in ChangChun,and proved that this model played a decision-supporting role of urban rail transit financialsubsidies.
(3) By analyzing real-time scheduling decision of urban rail transit, this paper studiedthe essence of real-time scheduling decision and summarized the lack of existing data,methods and techniques.then pointed out the clarity of passengers travel destination pointand TOD to improve the integrity of the original data. New Data collection method andimprovement program to the ticketing system at this stage were given to collect morecomplete travel information of passengers and to achieve real-time scheduling. TOD wasobtained to calculate the correction factor of passenger data by combining OD in thehistorical data and the vehicle running time line according to the basic idea of informationfusion. The paper also obtained the one-way real-time non-uniform departure scheduleswhich would be the basis of vehicle scheduling by making model that aimed at maximizinggrid spacing and used the grid spacing between two cars as variables of the model,meanwhile considered minimum grid spacing, the the maximum waiting time whichpassenger can tolerate and constraint of carriage capacity.
(4) System type selection of urban rail transit by introducing the evidence theory could be defined as: by applying probability assignment of a variety of system typesdetermined in each city as evidence, and then combine the evidence by certain rules tomake the integration of urban rail transit system type selection decisions. Based on thereasoning method of the D-S evidence theory, type selection decisions of urban rail transitsystem was defined as the domain, the type of urban rail transit system was defined as theelements on the domain, indicators related to construction decisions of urban rail transitsystem was defined as evidence.Then the evidence was combined to draw the integratedprobability of support functions in order to do decision-making for urban rail transit systemtype selection. So the model of urban rail transit system type selection was made,combination rules which apply to urban rail transit system type selection were elaboratedand solved problems with combination method of evidence.
The research of this paper could enrich and expand the theoretical framework andmethodology of operation and development decision-making methods for urban rail transit,and provide the theoretical basis for urban rail transit operations and developmentdecision-making.
本文结合我国城市轨道交通的发展现状，分析城市轨道交通系统内各子系统的信息化程度以及信息的完整性，对城市轨道交通系统选型、运营调度方案和政策性亏损补贴等问题的理论和方法开展了探索性研究。研究期望得出一套科学的、系统化的城市轨道交通运营决策的理论和方法体系，确定城市轨道交通系统的补贴方式，明确财政补贴的额度；完善数据采集方法，进一步实行数据挖掘，提高数据的利用率，实现动态实时调度，提高城市轨道交通运输效率；明确城市轨道交通中各条线路的功能定位和性能标准，并选择与之相适应的系统型式，以满足城市发展和城市交通的要求，辅助各城市进行城市轨道交通的规划决策，并为国家管理决策部门的决策管理提供依据，保障我国城市轨道交通的可持续发展。研究中运用信息融合理论和技术方法对城市轨道交通系统选型、运营调度方案和政策性亏损补贴等问题进行了理论上的探索和决策方法的研究。具体研究内容和创新性成果如下：
(1)在综合分析信息融合理论和城市轨道交通信息化内容的基础上，提出了城市轨道交通信息融合理论体系，阐述了该理论体系的概念定义、核心理念、从属关系等内容；将城市轨道交通决策问题分为数据层融合、特征层融合和决策层融合三个层次，明确了各个层次决策的输入输出单元类型、决策目标以及需要研究解决的内容和相互间的逻辑从属关系，并针对各个层次给出了对应的实际决策问题，分别为基于数据融合的城市轨道交通财政补贴决策，基于特征融合的城市轨道交通实时调度决策以及基于决策融合的城市轨道交通系统型式选择决策。
(2)对城市轨道主要的技术指标进行了汇总介绍。将城市轨道交通财政补贴问题分解为财政补贴方式选择和财政补贴决策两个阶段。基于信息融合的理论和方法，将城市轨道交通各个系统直接采集到的数据进行相关和集成融合计算出客流量、平均发车间隔和平均运距等。以社会福利最大化为目标，以票价、发车间隔、站间距以及服务区域面积等为决策变量，对城市轨道交通财政补贴问题进行了优化。计算结果表明，在社会福利最大化的目标下，收支平衡财政补贴政策适合于城市轨道交通系统。在此基础上，通过对车辆运营成本与乘客等车时间成本的分析，确定了乘客总的等车时间成本是常量固定的社会成本。利用边际成本法建立了城市轨道交通补贴决策模型。最后结合长春市轻轨的运营数据进行了实证分析，经检验证明该模型对城市轨道交通财政补贴具有辅助决策作用。
(3)从城市轨道交通实时调度决策问题的分析入手，探寻调度决策问题的实质，进而总结出现有数据、方法和技术的不足。为完善原始数据的完整性，提出了乘客出行讫点明确性和TOD。探讨了新的数据采集方法，给出了对现阶段售票系统的改进方案，从而可以实时的采集到更完整的乘客出行信息，使实时调度的实现成为可能。基于信息融合理论的基本思想，将历史数据中的OD与车辆运行时间线相融合，得出TOD进而计算出客流数据的特征系数。并且以后车与前车的发车间隔作为模型的变量，以发车间隔最大为目标，同时考虑最小发车间隔、乘客所能容忍的最大等车时间和车厢容量的约束建立模型，得出单方向实时的非均匀的发车时刻表，作为车辆调度的依据。
(4)将证据理论引入城市轨道交通系统型式选择决策问题，把城市轨道交通系统型式选择问题定义为：以应用各个城市指标确定的各种系统型式的概率赋值作为证据，再将这些证据按照一定的组合规则进行组合做出城市轨道交通系统型式选择决策的融合问题。根据D-S证据理论的推理方法，将城市轨道交通系统型式选择决策定义为论域；城市轨道交通系统的型式定义为论域内的元素；与城市轨道交通系统建设决策有关的指标定义为证据；将各条证据进行组合得出城市轨道交通系统型式选择的综合概率支持函数，从而为城市轨道交通系统型式选择做出决策。据此，建立了城市轨道交通系统型式选择模型，探讨了适用于城市轨道交通系统型式选择问题的组合规则，并应用所得证据组合方法进行求解。
本文的研究成果将丰富与拓展城市轨道交通运营与发展决策方法的理论框架与方法体系，为城市轨道交通运营与发展决策提供理论依据。
With the acceleration of the urban social and economic development and urbanizationprocess, there is a sharp increase in urban population, a large number of population floatinto the city, Staff travel and commodity exchange occur frequently,so the situation of theurban traffic is becoming grim. Many large and medium cities of the world have theproblem of road congestion and traffic disorder, the contradiction of the urban traffic ismore and more prominent, which lead to restrict the development of urban economy. Manybig cities at home and abroad have pointed out the development model of urban passengertransport which based on the urban rail transit and public transport in order to address theproblem of urban traffic congestion. But many factors, such as funding, environment,decision-making and policy, are restricting the development of urban rail transit.So it hasimportant practical and theoretical value to study on the urban rail transit system selection,operation scheduling scheme and policy subsidies for losses.
Based on the current development of China’s urban rail transport, this paper analysedthe degree and the integrity of the information of each subsystem in the urban rail transitsystem, and carried out exploratory research on the theories and methods of urban railtransit system selection, operation scheduling scheme and Policy subsidies for losses. A setof theory and methodology of operational decision-making for urban rail transit which isscientific and systematic was expected to draw to ascertain the subsidies for urban railtransit system and the amount of financial subsidies, to improve the data collectionmethods which can do further data mining and improve the utilization of data, to achievedynamic real-time scheduling in order to improve the efficiency of urban rail transit, toclear the functional position and performance standards of urban rail transit lines and selectthe corresponding system type in order to meet the requirements of urban transportationand make planning decisions of urban rail transit in cities,and to provide the basis fordecision management of the state department. The paper made theoretical exploration anddecision-making method on urban rail transit system selection, Operation schedulingscheme and Policy subsidies for losses by using the theory and techniques of informationfusion. Specific research and innovation results are given in the following:
(1) Based on the the theory of information fusion and the information of urban rail transit, This paper pointed out the theoretical system for information fusion of urban railtransit and elaborated the conception,core concept and affiliation of the theoretical system;The information fusion of urban rail transit was divided into three levels: datafusion,feature fusion and decision fusion,then cleared the type of input and output unit foreach level decision-making,and each level has correspnding decision-making: financialsubsidy decisions of urban rail transit based on data fusion, real-time scheduling decisionsof urban rail transport based on feature fusion and type selection decisions of urban railtransit system based on decision fusion.
(2) The paper made summary introduction about the main technical indicators ofurban rail transit.data from each system of urban rail transit was collected and fused tocalculate the passenger traffic, the average grid spacing and average distance. Aimed atmaximizing the social welfare, the paper optimized the financial subsidies of urban railtransit according to the decision variables such as the fare, the grid spacing, station spacingand service area, etc. The calculation results showed that break-even financial subsidypolicy was suitable for urban rail transit system to the goal of maximizing social welfare.On this basis, it could be identified that the total waiting time cost of passengers wasconstant fixed social cost by analyzing vehicle operating cost and waiting time cost ofpassengers, and made the decision model of urban rail transit subsidy decision by usingmarginal cost method.Finally analysized the operational data of light rail in ChangChun,and proved that this model played a decision-supporting role of urban rail transit financialsubsidies.
(3) By analyzing real-time scheduling decision of urban rail transit, this paper studiedthe essence of real-time scheduling decision and summarized the lack of existing data,methods and techniques.then pointed out the clarity of passengers travel destination pointand TOD to improve the integrity of the original data. New Data collection method andimprovement program to the ticketing system at this stage were given to collect morecomplete travel information of passengers and to achieve real-time scheduling. TOD wasobtained to calculate the correction factor of passenger data by combining OD in thehistorical data and the vehicle running time line according to the basic idea of informationfusion. The paper also obtained the one-way real-time non-uniform departure scheduleswhich would be the basis of vehicle scheduling by making model that aimed at maximizinggrid spacing and used the grid spacing between two cars as variables of the model,meanwhile considered minimum grid spacing, the the maximum waiting time whichpassenger can tolerate and constraint of carriage capacity.
(4) System type selection of urban rail transit by introducing the evidence theory could be defined as: by applying probability assignment of a variety of system typesdetermined in each city as evidence, and then combine the evidence by certain rules tomake the integration of urban rail transit system type selection decisions. Based on thereasoning method of the D-S evidence theory, type selection decisions of urban rail transitsystem was defined as the domain, the type of urban rail transit system was defined as theelements on the domain, indicators related to construction decisions of urban rail transitsystem was defined as evidence.Then the evidence was combined to draw the integratedprobability of support functions in order to do decision-making for urban rail transit systemtype selection. So the model of urban rail transit system type selection was made,combination rules which apply to urban rail transit system type selection were elaboratedand solved problems with combination method of evidence.
The research of this paper could enrich and expand the theoretical framework andmethodology of operation and development decision-making methods for urban rail transit,and provide the theoretical basis for urban rail transit operations and developmentdecision-making.
引文
[1]严波.城市轨道交通运营组织优化研究[D].东南大学,2006.
[2]城市轨道交通概况[R].网络版.2010.
[3]黄荣.城市轨道交通网络化运营的组织方法及实施技术研究[D].北京交通大学,2010.
[4]曹守华.城市轨道交通乘客交通特性分析及建模[D].北京交通大学,2009.
[5]王红梅.城市轨道交通的运营管理研究[D].北京交通大学,2007.
[6]赵良杰.公共交通的最优定价理论与模型研究[D].重庆大学,2006.
[7]李为为.城市轨道交通调度指挥智能集成系统研究[D].北京交通大学,2006.
[8]曹阳.城市轨道交通实时调度方法研究[D].吉林大学,2009.
[9]袁博晖.城市轨道交通客流分担率预测及类型选择研究[D].西南交通大学,2004.
[10]孟祥定.绿色交通视角下城市轨道交通网络规划决策方法及应用[D].湖南大学,2007.
[11] John Pucher, Stefan Kurth. Verkehrsverbund: the success of regional public transportin Germany, Austria and Switzerland [J]. Transportation Policy,1995,2(4):279-291.
[12] Matt G. Karlaftis and Patrick McCarthy. Operating subsidies and performance inpublic transit: an empirical study [J]. Transportation Research Part A: Policy and Practice,1998,32(5):359-375.
[13] Peter Tisato. Service unreliability and bus subsidy [J]. Transportation Research Part A:Policy and Practice,1998,32(6):423-436.
[14] Douglas O. A. Osula. A procedure for estimating transit subsidization requirements fordeveloping countries [J]. Transportation Research Part A: Policy and Practice,1998,32(8):599-606.
[15] Jianling Li and Martin Wachs. The effects of federal transit subsidy policy oninvestment decisions: The case of San Francisco's Geary Corridor [J]. Transportation,2004,31(1):43-67.
[16] James Odeck and Abdulrahim Alkadi. The performance of subsidized urban and ruralpublic bus operators: Empirical evidence from Norway [J]. The Annals of RegionalScience,2004,38(3):413-431.
[17] De Witte. A, Macharis. C, Lannoy. P, Polain. C, Steenberghen. T, Van de Walle. S. Theimpact of ''free'' public transport: The case of Brussels [J]. Transportation Research Part A:Policy and Practice,2006,40(8):671-689.
[18] Peter Nelson, Andrew Baglino, Winston Harrington, Elena Safirova and AbramLipman. Transit in Washington, DC: Current benefits and optimal level of provision [J].Journal of Urban Economics,2007,62(2):231-251.
[19] Rainald Borck and Matthias Wrede. Commuting subsidies with two transport modes[J]. Journal of Urban Economics,2008,63(3):841-848
[20] Ying Zhou, Hong Sok Kim, Paul Schonfeld and Eungcheol Kim. Subsidies andwelfare maximization tradeoffs in bus transit systems [J]. The Annals of Regional Science,2008,42(3):643-660.
[21] Avishai Ceder. Bus frequency determination using passenger count data [J],Transportation Research Part A: General,1984,18(5-6):439-453.
[22] A. A. Bertossi, P. Carraresi, G. Gallo. On some matching problems arising in vehiclescheduling models[J], Networks,1987,17(3):271-281.
[23] Andrzej Adamski and Andrzej Turnau. Simulation support tool for real-timedispatching control in public transport [J]. Transportation Research Part A: Policy andPractice,1998,32(2):73-87.
[24] A. Ceder, B. Golany, O. Tal. Creating bus timetables with maximal synchronization[J].Transportation Research Part A: Policy and Practice,2001,35(10:)913-928.
[25] Paolo Delle Site and Francesco Filippi. Bus service optimisation with fuel savingobjective and various financial constraints [J]. Transportation Research Part A: Policy andPractice,2001,35(2):157-176.
[26] André de Palma and Robin Lindsey. Optimal timetables for public transportation [J].Transportation Research Part B: Methodological,2001,35(8):789-813.
[27] Ali Haghani and Mohamadreza Banihashemi. Heuristic approaches for solvinglarge-scale bus transit vehicle scheduling problem with route time constraints [J],Transportation Research Part A: Policy and Practice,2002,36(4):309-333.
[28] Paul Calamai, Liping Fu, Qing Liu. Real-Time Optimization Model for DynamicScheduling of Transit Operations [J]. Transportation Research Record,2003,1857:48-55
[29] Saeed Zolfaghari, Nader Azizi and Mohamad Y. Jaber. A model for holding strategy inpublic transit systems with real-time information [J]. International Journal of TransportManagement,2004,2(2):99-110
[30] Shangyao Yan and Ching-Hui Tang. Inter-city bus scheduling under variable marketshare and uncertain market demands [J]. Omega,2009,37(1):178-192.
[31] Smith, E. An economic comparison of urban railways and express bus services[J].Journal of Transport Economic and Policy,1973.7:20-31.
[32] Dewees, D. N. Urban express bus and railroad performance: some Torontosimulations[J]. Journal of Transport Economic and Policy,1976.10:16-25.
[33] Boyd, J. H., Asher, N. J.,&Wetzler, E. S. Nontechnological innovation in urban transit:a comparison of some alternative[J]. Journal of Urban Economics,1978.5:1-20.
[34] Allport, R. J. The costing of bus, light rail transit and metro public transportsystems[J]. Traffic Engineering and Control,1981.22:633-639.
[35] Polydoropoulou, A.&Ben-Akiva, M. Combined revealed and stated preferencenested logit access and mode choice model for multiple mass transit technologies[J].Transportation Research Record,2001.1771:38-45.
[36] Jara-Díaz, S. R.,&Gschwender, A. Towards a general microeconomic model for theoperation of public transport [J]. Transport Reviews,2003.23(4):453-469.
[37] Bruun, E. Bus rapid transit and light rail: comparing operating costs with a parametriccost model[J]. Transportation Research Record,2005.1927:11-21.
[38] Desaulniers, G.,&Hickman, M. D. Public transit. Handbooks in operations researchand management science[J]. In Transportation. Barnhart and Laporte,2007. Vol.14.Elsevier. pp.69-127.
[39] Litman, T. Evaluatingrail transit benefits: acomment[J]. Transport Policy,2007.14:98-102.
[40] Alejandro Tirachini, David A. Hensher, Sergio R. Jara-Díaz. Comparing operator andusers costs of light rail, heavy rail and bus rapid transit over a radial public transportnetwork[J]. Research in Transportation Economics,2010,29:231-242.
[41]杨则海.城市公共交通的二重性与补贴机制研究[J].城市公共交通,2000,(04):15-17.
[42]张敏,欧国立.城市公共交通补贴问题分析[J].城市公共交通,2001,(03):9-12.
[43]王健,安实,赵泽斌.基于财政补贴的拥挤定价下公交收费策略研究[J].管理工程学报,2006,20(02):84-88.
[44]周春燕,王琼辉.公众参与城市轨道交通政府补贴机制探讨[J].价格理论与实践,2007,(06):26-27.
[45]赵源,欧国立.城市轨道交通补贴机制研究[J].北京交通大学学报(社会科学版),2008,7(2):7-11.
[46]杨天薇.对城市轨道交通补贴方式的探讨[J].现代城市轨道交通,2008,6:67-70.
[47]张一帆.城市公共交通补贴效率研究[D].北京交通大学.2009.
[48]郝记秀,周伟,黄浩丰和关宏志.城市公共交通财政补贴测算模型研究[J].交通运输系统工程与信息,2009,9(2):11-16.
[49]张晶.北京城市轨道交通补贴机制研究[D].北京交通大学.2010.
[50]贾腾,欧国立.城市轨道交通补贴依据与方法[J].学理论,2010,19:89-91.
[51]徐向阳,徐大伟和马洪冲.公交调度中放车调度模型的研究[J].交通与计算机,2003,4:19-21.
[52]陈芳.城市公交调度模型研究[J].中南公路工程,2005,30(2):162-164.
[53]宋瑞,赵航.基于机会约束的公交调度研究[J].数学的实践与认识,2005,35(1):89-95.
[54]刘鑫,何世伟.基于随机相关机会规划的公交调度模型研究[J].交通标准化,2006,12:152-155.
[55]戴连贵,刘正东.公交调度发车间隔多目标组合优化模型[J].交通运输系统工程与信息,2007,7(4):43-46.
[56]任传祥,郇宜军和尹唱唱.基于遗传禁忌搜索算法的公交调度研究[J].山东科技大学学报（自然科学版）,2008,27(4):53-56.
[57]郑小花,陈淑燕和武林芝.模拟退火算法在公交调度中的应用[J].信息化研究,2009,35(9):45-47.
[58]王敏,王楷.基于粒子群优化的公交调度研究[J].信息技术,2009,12:111-113.
[59]杨磊,刘卫朋和周磊.基于改进的随机公交调度问题的数学模型[J].河北工业大学学报,2010,39(1):74-78.
[60]叶丽萍,张文峰.基于蚁群算法的大型活动智能公交调度[J].中国科技信息,2010,8:255-257.
[61]陆梁.城市轨道交通的发展、分类与系统选择[J].城市轨道交通研究,1999,(2):7-10.
[62]周翊民.城市轨道交通的系统选择及其装备的国产化[J].城市轨道交通研究,2001,(3):1-6.
[63]岳渠德,马超,张晓东.我国城市轨道交通系统的选择[J].石家庄铁道学院学报,2001,14(3):44-46.
[64]李海川.城市单轨交通系统的选用探讨[J].铁道机车车辆,2003,23(6):40-42.
[65]施栩,魏庆朝.日本城市轨道交通应用系统模式[J].都市快轨交通,2004,17(5):54-59.
[66]蒋雅君,杨其新.城市轨道交通系统的分类及选型[J].城市轨道交通研究,2005,(2):70-73.
[67]杨敏,王炜,陈学武等.基于DEA/AHP方法的大运量快速交通方式选择决策[J].公路交通科技,2006,23(7):111-115.
[68]龙许友,魏庆朝.城市轨道交通系统模式与选型[J].综合运输.2007,11:26-29.
[69]杨耀.国外大城市轨道交通市域线的发展及其启示[J].城市轨道交通研究,2008,2:17-21.
[70]孔令洋.城市轨道交通系统型式选择研究[D].北京交通大学.2009.
[71]娄策群,桂学文,赵云合.信息化管理理论与实践[M].北京：北京交通大学出版社.2010.
[72]肖南峰.企业信息化[M].北京：清华大学出版社.2011.
[73]谭跃进,黄金才,朱承.决策支持系统[M].北京：电子工业出版社.2011.
[74]傅湘玲,欧阳峰.信息化管理教程[M].北京：清华大学出版社.2010.
[75]王正友.企业信息融合及其相关技术[M].北京：中国财政经济出版社.2008.
[76]彭冬亮.多传感器多源信息融合理论及应用[M].北京：科学出版社.2010.
[77]韩崇昭,朱洪艳,段战胜等.多源信息融合[M]北京：清华大学出版社.2010.
[78]杨国胜,窦丽华.数据融合及其应用[M].北京：兵器工业出版社.2004.
[79]杨万海.多传感器数据融合及其应用[M].西安：西安电子科技大学出版.2004.
[80]刘同明,夏祖勋,解洪成.数据融合技术及其应用[M].北京：国防工业出版社.1998.
[81]王润生.信息融合[M].北京：科学出版社.2007.
[82]王志胜,姜斌,甄子洋.融合估计与融合控制[M].北京：科学出版社.2009.
[83]杨露菁,余华.多源信息融合理论与应用[M].北京：北京邮电大学出版社.2005.
[84]何友,王国宏,关欣.信息融合理论及应用[M].北京：电子工业出版社.2010.
[85]蔡自兴,徐光祐.人工智能及其应用[M].北京：清华大学出版社.2004.
[86]蔡君时.城市轨道交通[M].上海：同济大学出版社.2000.
[87]毛保华.城市轨道交通[M].北京：科学出版社.2001.
[88]孙章,何宗华,徐金祥.城市轨道交通概论[M].北京：中国铁道出版社.2000
[89]张戎,李枫.城市轨道交通企业管理[M].北京：中国铁道出版社.2000.
[90]应名洪.城市轨道交通网络化建设与运营[M].北京：中国铁道出版社.2007.
[91]毛保华.城市轨道交通系统运营管理[M].北京：人民交通出版社.2006.
[92]何友等.多传感器信息融合及其应用[M].北京：电子工业出版社.2000.
[93] Mohring Herbert. Optimization and scale economies in bus transportation [J],American Economic Review. Vol.62, No.4(Sep.,1972),591-604.
[94] Mohring Herbert. Transportation Economics [M]. Ballinger.1976.
[95] Dubois D, Prade H. Representation and combination of uncertainty with belieffunctions and possibility measures [J]. Computational Intelligence,1988,4:244-264.
[96] Yager R. On the relationships of methods of aggregation of evidence in expertssystems [J]. Cybernetics and Systems,1985,16:1-21.
[97] Takahiko. H, Decision rule for pattern classification by integrating interval featurevalues[J]. Pattern Analysis and Machine Intelligence,1998,20(4):440-447.
[98]孙全,叶秀清,顾伟康.一种新的基于证据理论的合成公式[J].电子学报,2000,28(8):117-119.
[99] Smets P. The combination of evidence in the transferable belief model [J]. IEEE Trans.PAMI,1990,12（5）:477-458.
[100] Lefevre E, Colot O, Vannnoorenberghe P. Belief functions combination and conflictmanagement [J]. Information Fusion,2002,3（2）:149-163.
[101] Lefevre E, Colot O, Vannnoorenberghe P. Reply to the comments of R. Haenni onthe paper: belief functions combination and conflict management [J]. Information Fusion,2003,4（1）:63-65.
[2]城市轨道交通概况[R].网络版.2010.
[3]黄荣.城市轨道交通网络化运营的组织方法及实施技术研究[D].北京交通大学,2010.
[4]曹守华.城市轨道交通乘客交通特性分析及建模[D].北京交通大学,2009.
[5]王红梅.城市轨道交通的运营管理研究[D].北京交通大学,2007.
[6]赵良杰.公共交通的最优定价理论与模型研究[D].重庆大学,2006.
[7]李为为.城市轨道交通调度指挥智能集成系统研究[D].北京交通大学,2006.
[8]曹阳.城市轨道交通实时调度方法研究[D].吉林大学,2009.
[9]袁博晖.城市轨道交通客流分担率预测及类型选择研究[D].西南交通大学,2004.
[10]孟祥定.绿色交通视角下城市轨道交通网络规划决策方法及应用[D].湖南大学,2007.
[11] John Pucher, Stefan Kurth. Verkehrsverbund: the success of regional public transportin Germany, Austria and Switzerland [J]. Transportation Policy,1995,2(4):279-291.
[12] Matt G. Karlaftis and Patrick McCarthy. Operating subsidies and performance inpublic transit: an empirical study [J]. Transportation Research Part A: Policy and Practice,1998,32(5):359-375.
[13] Peter Tisato. Service unreliability and bus subsidy [J]. Transportation Research Part A:Policy and Practice,1998,32(6):423-436.
[14] Douglas O. A. Osula. A procedure for estimating transit subsidization requirements fordeveloping countries [J]. Transportation Research Part A: Policy and Practice,1998,32(8):599-606.
[15] Jianling Li and Martin Wachs. The effects of federal transit subsidy policy oninvestment decisions: The case of San Francisco's Geary Corridor [J]. Transportation,2004,31(1):43-67.
[16] James Odeck and Abdulrahim Alkadi. The performance of subsidized urban and ruralpublic bus operators: Empirical evidence from Norway [J]. The Annals of RegionalScience,2004,38(3):413-431.
[17] De Witte. A, Macharis. C, Lannoy. P, Polain. C, Steenberghen. T, Van de Walle. S. Theimpact of ''free'' public transport: The case of Brussels [J]. Transportation Research Part A:Policy and Practice,2006,40(8):671-689.
[18] Peter Nelson, Andrew Baglino, Winston Harrington, Elena Safirova and AbramLipman. Transit in Washington, DC: Current benefits and optimal level of provision [J].Journal of Urban Economics,2007,62(2):231-251.
[19] Rainald Borck and Matthias Wrede. Commuting subsidies with two transport modes[J]. Journal of Urban Economics,2008,63(3):841-848
[20] Ying Zhou, Hong Sok Kim, Paul Schonfeld and Eungcheol Kim. Subsidies andwelfare maximization tradeoffs in bus transit systems [J]. The Annals of Regional Science,2008,42(3):643-660.
[21] Avishai Ceder. Bus frequency determination using passenger count data [J],Transportation Research Part A: General,1984,18(5-6):439-453.
[22] A. A. Bertossi, P. Carraresi, G. Gallo. On some matching problems arising in vehiclescheduling models[J], Networks,1987,17(3):271-281.
[23] Andrzej Adamski and Andrzej Turnau. Simulation support tool for real-timedispatching control in public transport [J]. Transportation Research Part A: Policy andPractice,1998,32(2):73-87.
[24] A. Ceder, B. Golany, O. Tal. Creating bus timetables with maximal synchronization[J].Transportation Research Part A: Policy and Practice,2001,35(10:)913-928.
[25] Paolo Delle Site and Francesco Filippi. Bus service optimisation with fuel savingobjective and various financial constraints [J]. Transportation Research Part A: Policy andPractice,2001,35(2):157-176.
[26] André de Palma and Robin Lindsey. Optimal timetables for public transportation [J].Transportation Research Part B: Methodological,2001,35(8):789-813.
[27] Ali Haghani and Mohamadreza Banihashemi. Heuristic approaches for solvinglarge-scale bus transit vehicle scheduling problem with route time constraints [J],Transportation Research Part A: Policy and Practice,2002,36(4):309-333.
[28] Paul Calamai, Liping Fu, Qing Liu. Real-Time Optimization Model for DynamicScheduling of Transit Operations [J]. Transportation Research Record,2003,1857:48-55
[29] Saeed Zolfaghari, Nader Azizi and Mohamad Y. Jaber. A model for holding strategy inpublic transit systems with real-time information [J]. International Journal of TransportManagement,2004,2(2):99-110
[30] Shangyao Yan and Ching-Hui Tang. Inter-city bus scheduling under variable marketshare and uncertain market demands [J]. Omega,2009,37(1):178-192.
[31] Smith, E. An economic comparison of urban railways and express bus services[J].Journal of Transport Economic and Policy,1973.7:20-31.
[32] Dewees, D. N. Urban express bus and railroad performance: some Torontosimulations[J]. Journal of Transport Economic and Policy,1976.10:16-25.
[33] Boyd, J. H., Asher, N. J.,&Wetzler, E. S. Nontechnological innovation in urban transit:a comparison of some alternative[J]. Journal of Urban Economics,1978.5:1-20.
[34] Allport, R. J. The costing of bus, light rail transit and metro public transportsystems[J]. Traffic Engineering and Control,1981.22:633-639.
[35] Polydoropoulou, A.&Ben-Akiva, M. Combined revealed and stated preferencenested logit access and mode choice model for multiple mass transit technologies[J].Transportation Research Record,2001.1771:38-45.
[36] Jara-Díaz, S. R.,&Gschwender, A. Towards a general microeconomic model for theoperation of public transport [J]. Transport Reviews,2003.23(4):453-469.
[37] Bruun, E. Bus rapid transit and light rail: comparing operating costs with a parametriccost model[J]. Transportation Research Record,2005.1927:11-21.
[38] Desaulniers, G.,&Hickman, M. D. Public transit. Handbooks in operations researchand management science[J]. In Transportation. Barnhart and Laporte,2007. Vol.14.Elsevier. pp.69-127.
[39] Litman, T. Evaluatingrail transit benefits: acomment[J]. Transport Policy,2007.14:98-102.
[40] Alejandro Tirachini, David A. Hensher, Sergio R. Jara-Díaz. Comparing operator andusers costs of light rail, heavy rail and bus rapid transit over a radial public transportnetwork[J]. Research in Transportation Economics,2010,29:231-242.
[41]杨则海.城市公共交通的二重性与补贴机制研究[J].城市公共交通,2000,(04):15-17.
[42]张敏,欧国立.城市公共交通补贴问题分析[J].城市公共交通,2001,(03):9-12.
[43]王健,安实,赵泽斌.基于财政补贴的拥挤定价下公交收费策略研究[J].管理工程学报,2006,20(02):84-88.
[44]周春燕,王琼辉.公众参与城市轨道交通政府补贴机制探讨[J].价格理论与实践,2007,(06):26-27.
[45]赵源,欧国立.城市轨道交通补贴机制研究[J].北京交通大学学报(社会科学版),2008,7(2):7-11.
[46]杨天薇.对城市轨道交通补贴方式的探讨[J].现代城市轨道交通,2008,6:67-70.
[47]张一帆.城市公共交通补贴效率研究[D].北京交通大学.2009.
[48]郝记秀,周伟,黄浩丰和关宏志.城市公共交通财政补贴测算模型研究[J].交通运输系统工程与信息,2009,9(2):11-16.
[49]张晶.北京城市轨道交通补贴机制研究[D].北京交通大学.2010.
[50]贾腾,欧国立.城市轨道交通补贴依据与方法[J].学理论,2010,19:89-91.
[51]徐向阳,徐大伟和马洪冲.公交调度中放车调度模型的研究[J].交通与计算机,2003,4:19-21.
[52]陈芳.城市公交调度模型研究[J].中南公路工程,2005,30(2):162-164.
[53]宋瑞,赵航.基于机会约束的公交调度研究[J].数学的实践与认识,2005,35(1):89-95.
[54]刘鑫,何世伟.基于随机相关机会规划的公交调度模型研究[J].交通标准化,2006,12:152-155.
[55]戴连贵,刘正东.公交调度发车间隔多目标组合优化模型[J].交通运输系统工程与信息,2007,7(4):43-46.
[56]任传祥,郇宜军和尹唱唱.基于遗传禁忌搜索算法的公交调度研究[J].山东科技大学学报（自然科学版）,2008,27(4):53-56.
[57]郑小花,陈淑燕和武林芝.模拟退火算法在公交调度中的应用[J].信息化研究,2009,35(9):45-47.
[58]王敏,王楷.基于粒子群优化的公交调度研究[J].信息技术,2009,12:111-113.
[59]杨磊,刘卫朋和周磊.基于改进的随机公交调度问题的数学模型[J].河北工业大学学报,2010,39(1):74-78.
[60]叶丽萍,张文峰.基于蚁群算法的大型活动智能公交调度[J].中国科技信息,2010,8:255-257.
[61]陆梁.城市轨道交通的发展、分类与系统选择[J].城市轨道交通研究,1999,(2):7-10.
[62]周翊民.城市轨道交通的系统选择及其装备的国产化[J].城市轨道交通研究,2001,(3):1-6.
[63]岳渠德,马超,张晓东.我国城市轨道交通系统的选择[J].石家庄铁道学院学报,2001,14(3):44-46.
[64]李海川.城市单轨交通系统的选用探讨[J].铁道机车车辆,2003,23(6):40-42.
[65]施栩,魏庆朝.日本城市轨道交通应用系统模式[J].都市快轨交通,2004,17(5):54-59.
[66]蒋雅君,杨其新.城市轨道交通系统的分类及选型[J].城市轨道交通研究,2005,(2):70-73.
[67]杨敏,王炜,陈学武等.基于DEA/AHP方法的大运量快速交通方式选择决策[J].公路交通科技,2006,23(7):111-115.
[68]龙许友,魏庆朝.城市轨道交通系统模式与选型[J].综合运输.2007,11:26-29.
[69]杨耀.国外大城市轨道交通市域线的发展及其启示[J].城市轨道交通研究,2008,2:17-21.
[70]孔令洋.城市轨道交通系统型式选择研究[D].北京交通大学.2009.
[71]娄策群,桂学文,赵云合.信息化管理理论与实践[M].北京：北京交通大学出版社.2010.
[72]肖南峰.企业信息化[M].北京：清华大学出版社.2011.
[73]谭跃进,黄金才,朱承.决策支持系统[M].北京：电子工业出版社.2011.
[74]傅湘玲,欧阳峰.信息化管理教程[M].北京：清华大学出版社.2010.
[75]王正友.企业信息融合及其相关技术[M].北京：中国财政经济出版社.2008.
[76]彭冬亮.多传感器多源信息融合理论及应用[M].北京：科学出版社.2010.
[77]韩崇昭,朱洪艳,段战胜等.多源信息融合[M]北京：清华大学出版社.2010.
[78]杨国胜,窦丽华.数据融合及其应用[M].北京：兵器工业出版社.2004.
[79]杨万海.多传感器数据融合及其应用[M].西安：西安电子科技大学出版.2004.
[80]刘同明,夏祖勋,解洪成.数据融合技术及其应用[M].北京：国防工业出版社.1998.
[81]王润生.信息融合[M].北京：科学出版社.2007.
[82]王志胜,姜斌,甄子洋.融合估计与融合控制[M].北京：科学出版社.2009.
[83]杨露菁,余华.多源信息融合理论与应用[M].北京：北京邮电大学出版社.2005.
[84]何友,王国宏,关欣.信息融合理论及应用[M].北京：电子工业出版社.2010.
[85]蔡自兴,徐光祐.人工智能及其应用[M].北京：清华大学出版社.2004.
[86]蔡君时.城市轨道交通[M].上海：同济大学出版社.2000.
[87]毛保华.城市轨道交通[M].北京：科学出版社.2001.
[88]孙章,何宗华,徐金祥.城市轨道交通概论[M].北京：中国铁道出版社.2000
[89]张戎,李枫.城市轨道交通企业管理[M].北京：中国铁道出版社.2000.
[90]应名洪.城市轨道交通网络化建设与运营[M].北京：中国铁道出版社.2007.
[91]毛保华.城市轨道交通系统运营管理[M].北京：人民交通出版社.2006.
[92]何友等.多传感器信息融合及其应用[M].北京：电子工业出版社.2000.
[93] Mohring Herbert. Optimization and scale economies in bus transportation [J],American Economic Review. Vol.62, No.4(Sep.,1972),591-604.
[94] Mohring Herbert. Transportation Economics [M]. Ballinger.1976.
[95] Dubois D, Prade H. Representation and combination of uncertainty with belieffunctions and possibility measures [J]. Computational Intelligence,1988,4:244-264.
[96] Yager R. On the relationships of methods of aggregation of evidence in expertssystems [J]. Cybernetics and Systems,1985,16:1-21.
[97] Takahiko. H, Decision rule for pattern classification by integrating interval featurevalues[J]. Pattern Analysis and Machine Intelligence,1998,20(4):440-447.
[98]孙全,叶秀清,顾伟康.一种新的基于证据理论的合成公式[J].电子学报,2000,28(8):117-119.
[99] Smets P. The combination of evidence in the transferable belief model [J]. IEEE Trans.PAMI,1990,12（5）:477-458.
[100] Lefevre E, Colot O, Vannnoorenberghe P. Belief functions combination and conflictmanagement [J]. Information Fusion,2002,3（2）:149-163.
[101] Lefevre E, Colot O, Vannnoorenberghe P. Reply to the comments of R. Haenni onthe paper: belief functions combination and conflict management [J]. Information Fusion,2003,4（1）:63-65.