基于运输情景的多式联运路径规划优化建模方法研究
发布时间：2018-09-14 18:35
【摘要】：国家经济社会的可持续发展需要高效的货物运输系统予以支撑。随着我国综合运输体系的不断完善,作为一种先进的运输组织形式,多式联运在运输实践中得到了极大的推广应用,多式联运服务网络日趋成熟,运输的经济效益和社会效益也得到了显著提升。多式联运服务网络的高效运行与管理离不开科学合理的运输规划。在多式联运服务网络“战略—战术(网络设计)—运营”三层规划体系中,运营层面中的路径优化直接面向客户多样化的运输需求,是多式联运服务网络经济性、时效性和可靠性等性能的直观体现,因而具有十分重要的研究意义与实践价值,也是运输规划领域的热点问题之一。而当前多式联运路径规划既有研究在优化建模和算法设计上存在的不足严重制约了其为实际问题提供决策支持的可行性。基于此,本文从系统工程的角度出发,综合考虑优化对象、运输服务模式、网络能力、商品流完整性、优化准则以及求解方法六项问题规划特征,从基础运输情景建模入手,根据货物性质(以普通集装箱为代表的常规货物和以危险品为代表的非常规货物)对多式联运路径规化问题进行细分,进一步系统研究了多式联运路径规划复杂运输情景建模及精确求解策略设计问题。本文的研究工作主要有下几个方面:(1)在问题规划方面,本文综合分析了多式联运路径规划问题的六项规划特征。尤其针对“运输服务模式设定”这一特征,本文在分析多式联运服务网络系统的基础上,重点研究了两类运输服务模式,即以公路运输服务为代表的灵活运营的服务模式和以铁路运输服务为代表基于开行方案的服务模式,进而归纳、总结了列车开行方案对多式联运路径规划的时空约束,从而将优化建模中的运输服务模式设定由单运输服务模式扩展到多运输服务模式上来,这也是论文的重要创新点之一。(2)在优化建模方面,本文首先研究了单商品流、单运输服务模式以及单目标优化等设定下的多式联运路径规划基础问题的优化建模,构建了0-1整数线性规划模型,并将该模型应用于多式联运路径规划对供需变动的敏感性分析上,根据定量计算分析,得到有助于供需双方规划多式联运路径,组织多式联运的若干运输策略。然后,基于基础运输情景问题的“点—弧”建模方法,本文以普通集装箱和危险品两类最具代表性的货物品类为研究对象,考虑多商品流/危险品流规划、多运输服务模式设定以及模糊列车载运能力设定,首先以多商品流广义费用最优化为目标,构建了复杂运输情景下普通集装箱多式联运路径规划问题的混合机会约束整数非线性规划模型,继而采用高斯烟羽模型和箱型模型对多式联运服务网络中的社会风险和环境风险进行了分析与度量,构建了环境风险阈值约束下的危险品多式联运路径规划问题的“广义费用—社会风险”双目标混合机会约束整数非线性规划模型。(3)在求解方法设计和案例优化分析方面,本文致力于精确求解策略的设计,也即首先识别非线性规划模型中的非线性部分,然后进行线性化转化,得到与之等价的线性规划模型,最后采用经典的精确求解算法(例如分支定界法)在数学规划软件(例如LINGO)中进行问题的仿真求解。而针对危险品多式联运路径规划双目标优化问题,本文采用集成模型线性化转化与标准化加权求和法的求解策略进行求解,得到双目标优化问题的帕累托边界。最后,本文以大规模实例为证,验证了两类复杂运输情景建模问题中优化模型与精确求解策略处理实际问题时的可行性,同时在优化结果分析中采用敏感性分析法和模糊模拟法对实例中多式联运路径规划优化结果与相关模型参数(包括模糊机会约束置信值和环境风险阈值)之间的关系进行了探讨,总结了多式联运路径规划优化结果随参数设定变动的变化规律。
[Abstract]:The sustainable development of national economy and society needs the support of efficient freight transportation system.With the continuous improvement of China's comprehensive transportation system,as an advanced form of transport organization,multimodal transport has been greatly promoted and applied in transportation practice.The network of multimodal transport service is becoming more and more mature,and the economic and social benefits of transportation are becoming more and more mature. Efficient operation and management of the multimodal transport service network can not be separated from scientific and rational transport planning. In the three-tier planning system of the multimodal transport service network "strategy-tactics (network design) -operation", the route optimization at the operational level is directly oriented to the diversified transport needs of customers, and it is a multimodal transport service. It is of great significance and practical value to study the economic, timeliness and reliability of service network, and it is also one of the hot issues in the field of transportation planning. Based on the feasibility of decision support, this paper, from the perspective of system engineering, comprehensively considers the six planning features of optimization object, transportation service mode, network capability, commodity flow integrity, optimization criteria and solving methods, starts with the basic transport scenario modeling, and according to the nature of goods (conventional goods represented by ordinary containers). And the non-conventional goods represented by dangerous goods are subdivided into several parts, and the modeling of complex transportation scenarios and the design of exact solution strategies for multimodal transport path planning are further studied systematically. Six planning features of the route planning problem. Especially in view of the characteristics of "transportation service mode setting", this paper focuses on two types of transportation service modes based on the analysis of the multimodal transport service network system, i.e. the flexible operation service mode represented by highway transportation service and the open service mode represented by railway transportation service. The service mode of the train operation scheme is summarized, and the space-time constraints of the train operation scheme on the multimodal transport path planning are summarized. Thus, the transportation service mode setting in the optimization modeling is extended from the single transport service mode to the multi-transport service mode. This is also one of the important innovations of this paper. (2) In the optimization modeling, this paper first introduces the optimization model. This paper studies the optimization modeling of the basic problems of multimodal transport path planning under the assumptions of single commodity flow, single transportation service mode and single objective optimization. A 0-1 integer linear programming model is constructed. The model is applied to the sensitivity analysis of multimodal transport path planning to the changes of supply and demand. According to the quantitative calculation and analysis, it is helpful to supply and demand. Then, based on the "point-arc" modeling method of the basic transport scenario problem, this paper takes the two most representative types of goods, ordinary containers and dangerous goods, as the research object, considering the multi-commodity/dangerous goods flow planning, multi-transportation service mode setting and so on. Firstly, a mixed chance-constrained integer nonlinear programming model for routing problem of general container multimodal transport under complex transport scenarios is constructed with the objective of generalized cost optimization of multi-commodity flow. Then, Gaussian plume model and box model are used to analyze the social risk in multimodal transport service network. The environmental risk is analyzed and measured, and the "generalized cost-social risk" dual-objective mixed chance-constrained integer nonlinear programming model for the path planning problem of multimodal transport of dangerous goods with environmental risk threshold constraints is constructed. (3) In the design of solution methods and case optimization analysis, this paper focuses on the exact solution strategy. Design, that is to say, first identify the nonlinear part of the nonlinear programming model, then linearize the transformation, get the equivalent linear programming model, and finally use the classical exact solution algorithm (such as branch and bound method) in the mathematical programming software (such as LINGO) to solve the problem simulation. In this paper, the Pareto boundary of the bi-objective optimization problem is obtained by solving the bi-objective optimization problem with the strategy of linearization transformation of integrated model and standardized weighted summation. Finally, a large-scale example is given to verify the practical application of the optimization model and the exact solution strategy in two kinds of complex transportation scenario modeling problems. At the same time, sensitivity analysis method and fuzzy simulation method are used to analyze the feasibility of the problem, and the relationship between the optimal results of multimodal transport path planning and the parameters of the relevant model (including fuzzy chance constrained confidence value and environmental risk threshold) is discussed. The results of multimodal transport path planning optimization are summarized. Change rule of parameter setting.
【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：U116.2
[Abstract]:The sustainable development of national economy and society needs the support of efficient freight transportation system.With the continuous improvement of China's comprehensive transportation system,as an advanced form of transport organization,multimodal transport has been greatly promoted and applied in transportation practice.The network of multimodal transport service is becoming more and more mature,and the economic and social benefits of transportation are becoming more and more mature. Efficient operation and management of the multimodal transport service network can not be separated from scientific and rational transport planning. In the three-tier planning system of the multimodal transport service network "strategy-tactics (network design) -operation", the route optimization at the operational level is directly oriented to the diversified transport needs of customers, and it is a multimodal transport service. It is of great significance and practical value to study the economic, timeliness and reliability of service network, and it is also one of the hot issues in the field of transportation planning. Based on the feasibility of decision support, this paper, from the perspective of system engineering, comprehensively considers the six planning features of optimization object, transportation service mode, network capability, commodity flow integrity, optimization criteria and solving methods, starts with the basic transport scenario modeling, and according to the nature of goods (conventional goods represented by ordinary containers). And the non-conventional goods represented by dangerous goods are subdivided into several parts, and the modeling of complex transportation scenarios and the design of exact solution strategies for multimodal transport path planning are further studied systematically. Six planning features of the route planning problem. Especially in view of the characteristics of "transportation service mode setting", this paper focuses on two types of transportation service modes based on the analysis of the multimodal transport service network system, i.e. the flexible operation service mode represented by highway transportation service and the open service mode represented by railway transportation service. The service mode of the train operation scheme is summarized, and the space-time constraints of the train operation scheme on the multimodal transport path planning are summarized. Thus, the transportation service mode setting in the optimization modeling is extended from the single transport service mode to the multi-transport service mode. This is also one of the important innovations of this paper. (2) In the optimization modeling, this paper first introduces the optimization model. This paper studies the optimization modeling of the basic problems of multimodal transport path planning under the assumptions of single commodity flow, single transportation service mode and single objective optimization. A 0-1 integer linear programming model is constructed. The model is applied to the sensitivity analysis of multimodal transport path planning to the changes of supply and demand. According to the quantitative calculation and analysis, it is helpful to supply and demand. Then, based on the "point-arc" modeling method of the basic transport scenario problem, this paper takes the two most representative types of goods, ordinary containers and dangerous goods, as the research object, considering the multi-commodity/dangerous goods flow planning, multi-transportation service mode setting and so on. Firstly, a mixed chance-constrained integer nonlinear programming model for routing problem of general container multimodal transport under complex transport scenarios is constructed with the objective of generalized cost optimization of multi-commodity flow. Then, Gaussian plume model and box model are used to analyze the social risk in multimodal transport service network. The environmental risk is analyzed and measured, and the "generalized cost-social risk" dual-objective mixed chance-constrained integer nonlinear programming model for the path planning problem of multimodal transport of dangerous goods with environmental risk threshold constraints is constructed. (3) In the design of solution methods and case optimization analysis, this paper focuses on the exact solution strategy. Design, that is to say, first identify the nonlinear part of the nonlinear programming model, then linearize the transformation, get the equivalent linear programming model, and finally use the classical exact solution algorithm (such as branch and bound method) in the mathematical programming software (such as LINGO) to solve the problem simulation. In this paper, the Pareto boundary of the bi-objective optimization problem is obtained by solving the bi-objective optimization problem with the strategy of linearization transformation of integrated model and standardized weighted summation. Finally, a large-scale example is given to verify the practical application of the optimization model and the exact solution strategy in two kinds of complex transportation scenario modeling problems. At the same time, sensitivity analysis method and fuzzy simulation method are used to analyze the feasibility of the problem, and the relationship between the optimal results of multimodal transport path planning and the parameters of the relevant model (including fuzzy chance constrained confidence value and environmental risk threshold) is discussed. The results of multimodal transport path planning optimization are summarized. Change rule of parameter setting.
【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：U116.2
【参考文献】
相关期刊论文 前10条
1 张锦荣;李平;王文璇;;液氯运输泄漏事故扩散风险分析[J];中国安全生产科学技术;2015年07期
2 林枫;;基于Martins算法的联合运输最优路径规划[J];西南交通大学学报;2015年03期
3 黄丽霞;帅斌;;危险货物多式联运路径优化问题的多目标优化算法[J];中国安全生产科学技术;2014年09期
4 李愈;赵军;吴刚;陈佳琪;;带有固定运费的多式联运方式选择[J];西南交通大学学报;2012年05期
5 赵普生;徐晓峰;孟伟;董t，
本文编号：2243528
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/2243528.html
