【作者】 刘杰；

【导师】 陈旭梅；

【作者基本信息】 北京交通大学 ， 交通运输规划与管理， 2022， 硕士

【摘要】 随着新发展理念与“双碳”战略的提出,作为交通运输业碳排放贡献显著的货物运输行业,在推动社会转向更为可持续发展的过程中将发挥积极作用。因此,在交通减排社会责任及强制碳减排政策的双重驱动下,货物运输企业如何在满足运输需求的基础上实现成本与排放的均衡协调成为重要问题。为此,本文充分考虑碳减排可行政策,基于具有规模效应的轴辐运输模式,结合具有减排潜力的多式联运策略,针对货物运输企业进行考虑环境效益的多式联运轴辐网络设计与优化研究,研究成果对降低运输企业经济成本与排放、推动运输结构调整、构建低碳型运输服务体系具有重要意义。基于此,本文主要工作如下:首先,从考虑环境效益的低碳多式联运轴辐网络设计与优化模型构建,以及多式联运轴辐网络设计与优化模型求解算法等对两方面相关文献进行分析、总结,为本文考虑环境效益的多式联运轴辐网络模型构建与求解提供理论基础。其次,对多式联运轴辐网络的定义及特性进行分析总结,并对碳减排政策、规模效应、运输方式选择、经济成本等多式联运轴辐网络设计与构建影响因素进行解构与量化,强化本文构建的考虑环境效益的多式联运轴辐网络设计与优化模型对真实世界的刻画能力,提升本文研究的应用价值。随后,结合已确定的不同运输过程关键因素及量化特征,构建基于运输径路的成本测算模型和基于网络节点的碳排放量测算模型,并对不同阶段碳减排政策的特点与优势进行剖析,建立不同碳减排政策下考虑环境效益的多式联运轴辐网络设计与优化模型。在强制碳约束政策下,构建以经济成本和碳排放量最小为目标的社会责任导向型多目标多式联运轴辐网络设计与优化模型;在灵活碳交易政策下,耦合经济成本和碳交易成本与收益,构建以综合经济成本最小为目标的效益导向型单目标多式联运轴辐网络设计与优化模型,为实际应用提供理论支撑。然后,通过构建多式联运轴辐网络枢纽竞争力评价指标体系,确定枢纽选址。基于此,对多目标模型,提出归一化拥挤度计算方法改进NSGA-II算法求取初始可行解,为提高帕累托解集精度,提出基于等分法的初始聚类中心确定法改进KMeans算法对初始可行解进行聚类,最后融合ILS算法对可行解聚类代表进行再优化。对单目标模型,为避免后期解突变跳出最优解,提出使用非一致性变异改进遗传算法进行求解。最后,以河南省W货物运输公司为例,应用本文提出的模型,求解出不同的网络优化方案,并对其网络结构、网络成本、网络碳排放量进行比较分析。同时分别对碳交易价格、枢纽数量等进行敏感性分析,验证了模型的准确性与稳定性。更多还原

【Abstract】 With the introduction of the new development concept and the carbon peaking and carbon neutrality goals,the freight transport industry,as a significant contributor to carbon emissions in the transport industry,will play an active role in promoting society to be more sustainable.Thus,driven by both the social responsibility of emission reduction and the mandatory carbon emission reduction policy,it is important for freight transport enterprises to achieve a balanced coordination between cost and emission on the basis of satisfying transportation demand.Therefore,this thesis takes the feasible policy of carbon emission reduction into account,based on the hub-and-spoke network transport mode with scale effect,combined with the multimodal transport strategy with emission reduction potential,and conducts research on the design and optimization of multimodal hub-and-spoke network for freight transport enterprises considering environmental benefits.The main work of this thesis is as follows.Firstly,this thesis reviews the literature at home and abroad from two aspects,namely,the low-carbon multimodal hub-and-spoke network design and optimization model considering environmental benefits,and the algorithms for solving singleobjective and multi-objective optimization models.The above work provides the theoretical basis for the development and solution of the multimodal hub-and-spoke network design and optimization model considering environmental benefits.Secondly,the definition and characteristics of multimodal hub-and-spoke network are introduced.It also deconstructs and quantifies the factors influencing the design and optimization of multimodal hub-and-spoke network,such as carbon emission reduction policy,scale effect,transportation mode selection,and network cost,to strengthen the ability of the multimodal hub-and-spoke network design and optimization model constructed in this thesis to portray the real world,which can enhance the application value of this research.Subsequently,the cost measurement model based on transportation paths and the carbon emission measurement model based on network nodes are constructed by combining the identified key factors and quantitative characteristics of different transportation processes.Based on the above models,multimodal hub-and-spoke network design and optimization models considering environmental benefits are constructed based on the demonstration of the characteristics of different carbon reduction policies.In the mandatory carbon emission reduction stage,a social responsibility-oriented multimodal hub-and-spoke network design and optimization model is constructed with the objective of minimizing economic cost and carbon emission.In the carbon trade stage,a singleobjective cost-oriented multimodal hub-and-spoke network design and optimization model is constructed with the objective of minimizing the comprehensive economic cost by coupling the economic cost,carbon trading cost,and benefit.The above models provide theoretical support for practical applications.Then,by constructing an index system for evaluating the competitiveness of hubs in multimodal hubs and spokes networks,the hub sites are identified.Based on the hub location results,for the multi-objective model,the normalized congestion calculation method is proposed to improve the NSGA-II algorithm to find the initial feasible solution,and then to improve the accuracy of the Pareto solution set.The initial clustering center determination method based on the equipartition method is proposed to improve the KMeans algorithm to cluster the feasible solution,Finally the ILS algorithm is coupled to re-optimize the feasible solution clustering representatives.In order to avoid late solution mutations,the genetic algorithm using the non-coherent mutation method is proposed to for solving the single-objective model.Finally,the model proposed in this thesis is applied to solve different network optimization schemes for W cargo transportation company in Henan,China,where the network structure,economic cost and carbon emission are compared and analyzed.The accuracy and stability of the model are also verified by the sensitivity analysis on carbon trading price and number of hubs,respectively.更多还原