【作者】 许旺土；

【导师】 何世伟；

【作者基本信息】 北京交通大学 ， 交通运输规划与管理， 2010， 博士

【摘要】 快捷货运服务网络的构建及其运输能力的确定作为综合运输体系下货运研究的重要问题,受到越来越多从事货运工作的管理人员和研究人员的关注。鉴于现有成果并未系统地研究综合运输体系下以“快捷化”运输理念为主导的货运服务网络运输能力相关问题,本论文着重对综合运输体系下快捷货运(本论文简称为综合快捷货运)服务网络运输能力计算模型及其算法进行了研究,所做的工作主要包括以下七个部分：1.围绕综合快捷货运的发展趋势,阐述了综合快捷货运服务网络运输能力确定相关问题研究现状,分析了综合快捷货运服务网络运输能力计算的必要性。2.提出了综合货运物理网络和服务网络的概念,以及综合快捷货运服务网络的类型,以此为基础提出综合快捷货运服务网络运输能力的概念及构成,结合综合快捷货运服务网络运输能力构成部分的具体定义,分析了综合快捷货运服务网络运输能力的影响因素。3.分析了长途综合快捷货运服务网络运输能力计算方法,给出长途综合快捷货运服务网络的表示方法及其枢纽节点内部转运联弧处理方法,将交通工程学上的OD矩阵推算方法应用到长途综合快捷货运服务网络运输能力计算中,提出基于关键联弧装载单元流量与其运输能力偏差值最小的极大似然估计算法,确定了快捷货运量OD矩阵的增广系数,进而得到综合快捷货运服务网络的运输能力OD矩阵,采用算例对所提出的模型算法进行了验证。4.分析了区域综合快捷货运服务网络运输能力计算方法,给出时效性区域综合快捷货运服务网络的表示方法,提出了区域综合快捷货运服务网络上某OD对间K最短时空路径搜寻的“两阶段标号法”,结合改进Logit模型的时空路径快捷货物流量分配算法,提出基于K最短时空路径的时效性区域综合快捷货运服务网络运输能力的计算模型及算法,采用算例对所提出的模型算法进行了验证。5.确定了面向不同决策主体的综合快捷货运服务网络的运输能力。针对货运策略制订者,研究给出了不同运达时间限制条件下综合快捷货运服务网络最大有效运输能力、潜在运输能力的计算方法；针对货主和货运商,从货主选择货运商、货运商制订自身运输策略(基于最小运输费用的路径流量分配)角度出发建立综合快捷货运服务网络经济运输能力计算的双层规划模型,上层为货主选择货运商的快捷货运需求量平衡分配不等式,下层为以货运商、第三方承运人运输费用最小为目标的快捷货物流量分配模型,并采用算例对所提出的模型算法进行了验证。6.研究了考虑运达时间可靠性约束的综合快捷货运服务网络运输能力计算方法。针对快捷货物运输过程中,服务路段载运单元行程时间、快捷货物停留等待时间、枢纽节点内快捷货物转运作业时间为随机值的情况,提出带随机变量和约束条件的综合快捷货运服务网络的运输能力计算模型。设计了一个混合智能求解算法——在遗传算法中嵌入蒙特卡洛模拟法,结合算例验证了算法的效率并得出相应结论。7.采用中铁快运公铁联运的行邮服务网络进行实例验证,体现研究的实用价值及意义。更多还原

【Abstract】 As the key research topic of multimodal transportation system, structure and shipping capacity of express shipment service network have attached many attentions of both the mangers and the researchers of the multimodal commodity transportation. For the reason that there are no systematic researches on the shipping capacity of multimodal service network for express shipment under the idea of "express shipment", this thesis works on the computational techniques of multimodal service network capacity for express shipment. The main work includes the following interrelate parts:1. Along with the trend of multimodal express shipment, current situation as well as the necessity of computing capacity of multimodal service network for express shipment are reviewed and depicted.2. Concepts and classifies of multimodal service network for express shipment are proposed. Based on them, detail concepts as well as the constitution of the multimodal service network capacity for express shipment are put forward, including the analysis of the impact factor and simple measures of the service network capacity.3. Computational technique of multimodal long-haul service network capacity for express shipment is analyzed. Topology representation and transshipment link in hub point of multimodal long-haul service network for express shipment are put forward firstly. Then, with OD matrix estimation method in Traffic Engineering, the multimodal long-haul service network capacity matrix is determined. During that, the Maximum Likelihood Estimation on the difference between the container flow and its carrying capacity on the key service link are used to determine the increase factor of express shipment matrix which is used to calculate the capacity matrix. A numerical case is used to verify the proposed mathematical model and the solution algorithm.4. Computational technique of multimodal regional service network capacity for express shipment is analyzed. Topology representation of multimodal time-defined regional service network for express shipment is given out. In this process, the concepts, i.e. the physical route and the time-space path on multimodal time-defined regional service network are addressed. Based on the concepts of physical route and time-space path, a "two stages labeling" algorithm is proposed to search the K shortest time-space path between an OD pair. As the K shortest time-space path algorithm has been proposed, the computational method of multimodal time-defined regional service network capacity for express shipment is formulated as well. A numerical case is used to verify the proposed model and the solution algorithm.5. Computational techniques of multimodal service network capacity for express shipments with various decision individuals are studied. For policy constituter of commodity shipment, the maximal effective capacity and potential capacity of multimodal service network for express shipments versus different defined delivery time are determined; for the shipper and the carrier, standing on the point that the shippers choosing the carriers, while the carriers making up their own shipping strategies, the mathematical models for economic capacity of multimodal service network for express shipments are formulated, these two models are integrated with a bi-level programming. The upper programming cares about the process of the shippers’ choosing the carriers. On the other hand, the lower programming is the express shipment flow assignment model on the physical routes of the carriers and the logistics service provider (LSP). A numerical case is used to verify the proposed mathematical model and the solution algorithm.6. Computational technique of multimodal service network capacity for express shipments with reliability constraints of defined delivery time is researched. For the reason that during the commodity shipping process, the running time on service link of carrying unit, waiting time, transfer time in hub point of express commodities are all uncertain. The mathematical model for multimodal service network capacity computation for express shipments with stochastic variables and constraints is formulated. A hybrid intelligent algorithm (HIA) in which Monte Carlo simulation is combined with genetic algorithm (GA) is put forward to solve the proposed model. A numerical case is used to verify the conclusions and the efficiency of solution algorithm.7. Service network of highway-railway transportation for luggage and mail shipping of China Railway Co., Ltd. is taken as a real example to verify the studies in this thesis, which embodies application value and significance of this thesis.更多还原