【作者】 沈瑞光；

【导师】 裴玉龙；

【作者基本信息】 哈尔滨工业大学 ， 交通运输规划与管理， 2013， 博士

【摘要】 新时期交通建设的任务迫切地需要为人民生产和生活提供一套快速、舒适、便捷、安全的对外运输和城市交通服务系统，推动我国交通运输结构健康、有序、可持续的优化调整。近年来，我国以高速铁路为建设契机的综合交通枢纽成为对外交通与城市交通客流集散和中转的重要结点，2008年正式运行的北京南站，2010年启用的上海虹桥综合交通枢纽，2012年投入使用的哈尔滨哈西综合交通枢纽等等，它们成为了新时期交通运输发展的标志性基础设施。枢纽的客运需求直接影响到设施规模大小和枢纽工程造价，关乎到能否满足居民高品质的出行需求及枢纽工程的经济效益和抗风险能力。因此，如何合理准确地确定综合交通枢纽各方面的客运需求，成为控制综合交通枢纽工程规模、提高枢纽各方式服务水平的一个先导性课题。从交通工程学的角度界定城市综合客运枢纽的内涵，分析城市综合客运枢纽的特征，对综合客运枢纽的类型进行划分，结合城市类型，分析不同类型的城市所使用的枢纽类型。以旅客有准备、愿意、并有能力购买客运服务的需要为前提，界定城市综合客运枢纽客运需求研究范围，即枢纽对外客运需求、枢纽换乘需求及枢纽停车需求三个方面。在综合交通枢纽客运需求及客运特征调查与数据采集的基础上，分析旅客客运偏好意愿形成机理和选择规律，引入旅客对客运服务特征依存程度的概念，对客运意愿进行量化表征，即是旅客依存度对客运服务特征的映射。以不同客运服务的需求分配概率服从多项logit概率分布模型为假设前提，求解不同客运服务的客运意愿及依存度。在借鉴现有枢纽客运需求预测理论与方法的基础上，提出综合交通枢纽对外客运需求分析需按照“城市全社会对外客运需求预测—多方式客运分配需求预测—单枢纽单方式需求预测—单枢纽多方式需求预测”多层预测体系思路进行。在城市全社会背景下，确定4大类24个指标作为影响对外客运需求变化的关联元素作为输入变量，提出基于点积-平移型支持向量机的客运表征指标预测模型。选取可能对旅客客运意愿方式特征依存度产生影响的人均可支配收入、人均消费性支出等指标作为扰动变量，提出基于旅客依存度和对外客运服务特征的多方式序贯预测模型，计算未来一段时间旅客对不同对外客运方式需求的概率分布。以现状枢纽适站量、方向系数等数据调查为基础，预测未来年份特定枢纽对外客运发送量。预测结果表明，随着经济的发展和交通设施设备水平的提升，未来航空、高铁等高速化、舒适化的客运方式在全社会对外客运发展中的重要地位将有显著提升。明确综合交通枢纽换乘需求类别可以划分为对外交通方式间的换乘和对外交通与城市交通间的换乘两种。提出枢纽中两种对外客运方式换乘需求预测方法，构建以可达性控制因子和方便性因子为参数的高铁换乘公路的客运需求计算模型，并提出参数因子的标定方法。以旅客最小行动换取最大效用为假设前提，同时在假定枢纽旅客年龄结构不随时间改变的前提下，构建对外交通与城市交通间换乘预测模型，预测未来年份枢纽旅客对换乘方式的需求意愿和概率。从综合交通枢纽的影响范围和需求分析入手，将枢纽的停车需求划分为对外换乘停车需求、接送客停车需求、枢纽访客停车需求、转移停车需求四个部分。针对对外换乘停车需求的影响因素，构建S曲线需求预测模型，进行参数标定。提出基于不同需求类别的停车需求预测方法，以共享最优停车泊位利用为目标，提出综合交通枢纽共享式停车泊位需求计算方法，并通过案例分析对提出的预测模型进行验证。本文以研究城市综合交通枢纽客运需求预测方法与模型为突破点，提出一套服务于综合交通枢纽对外客运、换乘、停车三方面需求的预测方法与模型。能够为综合交通枢纽，特别是以面向高速铁路和城市轨道交通为主的综合交通枢纽的规划、可行性研究及设计工作提供借鉴和参考，对于改进完善综合交通枢纽客运需求预测方法、提高预测精度具有重要而积极的促进作用。更多还原

【Abstract】 The task of traffic construction in the new period urgently needed to provide afast, comfortable, convenient, safe transportation and urban transport service systemfor people’s production and life, which promote the optimization and adjustment ofChina’s transportation structure healthily, orderly and sustainablely. In recent years,by building high-speed railway, the comprehensive passenger transport hubs haveturned into the important distribution and transit junction in the externaltransportation and urban transport. Beijing South Railway Station operated in2008,hanghai Hongqiao integrated transport hub opened in2010, HaXi integratedtransport hub put into use in2012, and so on, which has become the symbol oftransport infrastructure of transport development in the new period. The passengertransport demand of hub directly affect the scale of the facility and project cost，related to meet travel demand and project residents high quality economic efficiencyand the ability to resist risks. Therefore, how to fix the sorts of passenger transportdemand of the comprehensive hub reasonably and accurately，become a guidingsubject to control the scale of hub and to improve the service level of the hub.Defining the connotation of urban integrated passenger transport hub from thepoint of transport engineering, the characteristics of urban integrated passengertransport hub have been analyzed, the suitablity of hub type have been analyzed,combined with the city type. The prerequisite is that the passengers have ready,willing, and need the ability to purchase passenger service. Define the studyingscope of passenger demand of urban integrated transport hub, which is referring tothree aspects of external passenger transport demand, transfer demand, and parkingdemand. Based on the investigation and analysis of the travelers’ characteristics insome hub, the formation mechanism of the passengers’ transport willingness andregular pattern of choice has been analyzed, the concept of the passengers’dependence degree on the traits of the passenger transport service has beenintroduced, passenger transport willingness has been characterized, which would bereflected by quantifying the passengers’ transport will. Taking the model thatdifferent probability distribution complying with multiple probability distribution ashypothesis, it would find out different kinds of transport desire and the related dependence degree.Drawing lessons from the existing forecast theory and method of passengertransport demand of the hub, this article brings forward that the analysis of the areatransportation demand should follow the multi-layer forecast system---“urbansociety’s external passenger transportation demand forecast---multi-modaltransportation demand distribution forecast---single hub single-modal demandforecast---single hub multi-modal demand forecast”. Against the urban societybackground,24indicators of four categories which are relevant factors influencingthe transportation demand would be taken as input variables, and the forecastmethod “dot product-translation vector machine” will be raised. Moreover, the otherindicators such as per-capita disposable income and per-capita consumptionexpenditure, will be taken as disturbance variables, and then predict the probabilitydistribution of the transportation modes in next period of time. It could be predictedthe passenger throughput under specific mode from the specific hub in the future,according to the current hub amount and the direction coefficient data survey. Thepredicted results show that with the development of economy and traffic facilitiesand equipment level, passenger transportation development of aviation, high-speedrail will be significantly improved based on their high-speed and comfortcharacteristics.Define the transferrfing demand of hub could be divided into two classes,namely between external transportation modes, and between external transportionand urban transport. Put forward the transferring demand forecasting approach ofbetween two kinds of external transportation modes. Taken two parameters ofaccessibility factors and convenience factors as evaluation indexes, a calculationmodel on passenger transfer demand between high speed railway and highway waspresent. The parameters calibration has been put forward. Take the passengerminimum action for maximum utility assumption, at the same time, under theassumption of passenger hub age structure does not change with time, constructionof transportation and urban transport transfer prediction model, predict the futureyear hub passenger on the transfer mode demands and probability.In view of the hub influence scope and passenger transfer demands, four typesof the parking demand was divided, namely, external-transfer, shuttle guest, tempvisitors and convert. By the influence factors of external-transfer parking demand, a forecast demand of S shape curve was used to evaluate the parameters. An approachis present to predict the different demand of parking. Taken sharing the optimalparking location as target function, the approach can be used to calculate the sharedparking location in the comprehensive hub. A case was analyzed and simulated toverify the proposed approach.It is the breaking point that is researching on the demand forecasting methodsand models. This article aims to put forward one set of analysis and forecast method,serving external transport, transferring and parking in the comprehensive hub. Thismethod system would provide some reference value for the planning, feasibilitystudy and design works of the integrated hub, particularly for the one mainly withhigh-speed rail and urban rail. It also would be helpful to improve the forecastmethod and to enhance the forecast precision.更多还原