以低碳为目标的集装箱接驳运输问题的建模与优化

发布时间：2018-08-06 15:27

【摘要】：联合国政府间气候变化委员会的一份报告指出：过去50年里全球平均气温的升高主要是排入大气的温室气体造成的,而全球变暖又将引发海平面升高、极端天气频发等恶劣自然现象,从而进一步影响人类的生产、生活甚至威胁人类的生存。物流运输业是二氧化碳等温室气体的重要来源之一,而道路货运的碳排放量在物流运输中占有重要的比例。目前,集装箱是货运的主要形式,所以集装箱道路运输是道路货运的重要组成部分。因此研究以低碳为目标的集装箱道路运输问题,可带来可观的环境效益。集装箱接驳运输是指在长距离的远洋或铁路运输开始之前或结束之后的距离相对较短的通常由卡车实现的运输环节。这种运输方式能够提供门到门的运输服务。虽然其运输的距离较短,但单位距离单位集装箱的运输成本却很高。此外这种运输方式容易造成道路拥堵,货物延迟交付和损坏等。因此,研究提高集装箱接驳运输的运输效率非常重要。本文对一个集装箱接驳运输问题进行了研究。不但考虑了运输任务的双时间窗(取货地和送货地)属性和空集装箱的调度,还考虑了接驳运输对环境的影响。以接驳运输过程中车辆总的碳排放量代替传统的以车辆总的行驶时间或运行成本为优化目标,并且对其进行了分析。本论文主要做了以下几个方面的研究工作：(1)对车辆行驶速度固定的集装箱接驳运输问题进行了研究。基于确定的活动在顶点上的图(DAOV图),建立了以低碳为目标的集装箱接驳运输问题的混合整数线性规划模型。测试了基于商业优化软件LINGO求解该模型的速度与精度,与传统的以总工作时间为目标的模型进行了对比,分析了时间窗宽度、进出口箱所占比例以及空箱所占比例对结果的影响。结果表明：这种方法能够在很短的时间内求得实际规模问题的高精度解,且可以有效降低碳排放量。(2)对车辆行驶速度未知的集装箱接驳运输问题进行了研究。对DAOV图进行了扩展,基于该图以车辆行驶速度为决策变量,建立了一个混合整数非线性规划模型。针对该模型设计了基于速度离散化的求解算法,并将该模型转化为混合整数线性规划模型。实验结果表明：这种基于速度离散化的算法能够在很短的时间内给出实际规模问题的近似最优解,对于能够通过其它方法给出理论最优解的中小规模算例,这种近似最优解与理论最优解的差距很小(一般小于0.1%)。与给定车辆行驶速度情形的对比进一步验证了这种模型和算法的有效性。(3)对车辆行驶速度未知的集装箱接驳运输问题进行了研究。基于DAOV图进行了扩展,以车辆的行驶时间为决策变量,建立了一个混合整数非线性规划模型。针对该模型设计了基于时间窗离散化的求解算法,并将模型转化为纯整数(0-1)线性规划模型。实验结果表明：这种基于时间窗离散化的算法能够在很短的时间内给出实际规模问题的近似最优解,对于能够通过其它方法给出理论最优解的中小规模算例,这种近似最优解与理论最优解的差距很小(一般小于0.1%)。与速度离散化算法的对比证明了该算法的有效性。
[Abstract]:A report by the United Nations Intergovernmental climate change Commission (Intergovernmental climate change Commission) says that the increase in global average temperature in the past 50 years is mainly caused by atmospheric greenhouse gases, and global warming will also cause severe natural phenomena such as rising sea levels, extreme weather and so on, which further affect human production, life and even threaten human beings. The logistics transportation industry is one of the important sources of carbon dioxide and other greenhouse gases, and the carbon emission of the road freight occupies an important proportion in the logistics transportation. At present, the container is the main form of the freight transportation, so the container road transportation is an important part of the road freight. Transport problems can bring considerable environmental benefits. Container transport is a transport link usually achieved by a truck before or after the start or end of a long distance ocean or railway transportation. This mode of transport provides a door to door transport service. Although the distance is short, the unit distance is single. The transport cost of the container is very high. In addition, it is easy to cause road congestion, delay in delivery and damage of goods. Therefore, it is very important to study the transport efficiency of container transport. In order to connect the total carbon emissions of vehicles in the transportation process instead of the traditional vehicle running time or operation cost, the optimization objective is to replace the traditional vehicle running time or operation cost. The main research work of this paper is as follows: (1) the vehicle The problem of container transport with fixed speed is studied. Based on the graph (DAOV) on the vertex, a mixed integer linear programming model is established for the container transport problem with the target of low carbon. The speed and accuracy of the model based on the commercial optimization software LINGO are tested and the traditional method is used. The effect of the time window width, the proportion of the inlet and outlet boxes and the proportion of empty boxes on the results is analyzed. The results show that this method can obtain the high precision solution of the actual scale problem in a very short time, and can effectively reduce the carbon emission. (2) the vehicle speed is unknown. The problem of container connection transportation is studied. The DAOV graph is extended. Based on the vehicle driving speed as a decision variable, a mixed integer nonlinear programming model is established. A solution algorithm based on velocity discretization is designed for the model. The model is transformed into a mixed integer linear programming model. Ming: the algorithm based on the velocity discretization can give an approximate optimal solution to the actual scale problem in a very short time. For a small and medium scale example that can give the optimal solution of the theory through other methods, the gap between the approximate optimal solution and the theoretical optimal solution is very small (generally less than 0.1%). The validity of this model and algorithm is further verified. (3) the problem of container transport with unknown vehicle speed is studied. Based on the DAOV graph, a model of mixed integer nonlinear programming is established with the driving time of the vehicle as a decision variable. The time window discretization is designed for the model. The algorithm is transformed into a pure integer (0-1) linear programming model. The experimental results show that the algorithm based on the time window discretization can give the approximate optimal solution of the actual scale problem in a very short time, and the approximate optimal solution of the approximate optimal solution for the best solution of the theory can be given by other methods. The difference between theoretical optimal solutions is very small (generally less than 0.1%). The effectiveness of the algorithm is demonstrated by comparison with the speed discretization algorithm.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U492.334

【相似文献】