中欧集装箱多式联运服务网络设计  

Design of China-Europe intermodal container transport service network

作　　者：艾子妍 张旭 武旭[1] AI Ziyan;ZHANG Xu;WU Xu(School of Traffic and Transportation,Beijing Jiaotong University,Beijing 100044,China;Civil Aviation Engineering Consulting Company of China,Beijing 100621,China)

机构地区：[1]北京交通大学交通运输学院,北京100044 [2]中国民航工程咨询有限公司,北京100621

出　　处：《铁道科学与工程学报》2024年第6期2217-2228,共12页Journal of Railway Science and Engineering

基　　金：中国国家铁路集团有限公司科技研究开发计划重点课题(N2023X035)。

摘　　要：中欧运输通道和运输方式不断发展完善,海运、中欧班列及多种运输方式联运等构成了中欧间集装箱运输服务网络。货主在选择运输服务时,一直关注货物运输的费用与时效性,由于中欧间运输距离长,节点多,节点作业时长还存在很大的不确定性。同时,随着全球对碳排放问题的重视,运输服务产生的碳排放也成为货主考虑的因素。综合考虑运输费用、时间和碳排放的影响,并关注节点作业时间的不确定性,解决中欧集装箱多式联运服务网络设计问题具有非常重要的现实意义。建立最小化运输费用、运输时间和运输碳排放量的多目标多式联运服务网络设计模型,并在模型中引入不确定性时间变量。由于节点作业时间样本数据有限,通过Box-Muller变换生成随机数丰富数据,并减少不可观测的误差,运用蒙特卡洛模拟对运输时间进行不确定性统计,描述总运输时间的统计特征。基于多目标的Pareto最优思想,设计了快速非支配排序遗传算法求解最优运输服务方案。以天津至汉堡的中欧集装箱运输为实例,根据实际调研结果确定各项相关参数设定,进行模型和算法验证,求解得到多式联运运输方案的Pareto最优解集。结果显示不同的运输服务方案其运输费用、运输时间、碳排放量各有差异,并且符合Pareto最优解集定义,证明了研究提出的考虑不确定性的多目标服务网络设计建模及算法的正确性和可行性,研究成果可为货主提供选择符合其需求的不同运输服务优化方案。With the improvement of the transport channels and modes between China and Europe,the container transport service network has been formed by maritime transport,China Railway Express,and intermodal transport.When choosing transport services,shippers always concern about the costs and transport timeliness.Due to the long distance and many nodes between China and Europe,there is a great uncertainty in the duration of node operation.Meanwhile,with the global emphasis on carbon emission issues,the carbon emission generated by transport services has also become a factor for shippers to consider.It is important to comprehensively consider the effects of transport cost,time and carbon emission,and pay attention to the uncertainty of node operation time to design the China-Europe container multimodal transport service network.A multi-objective intermodal transport service network design model was established to minimize transport cost,time and carbon emission,and uncertainty time variables were introduced into the model.Due to the limited sample data of node operation time,random numbers were generated by Box-Muller transformation to enrich the data and reduce the unobservable errors,and Monte Carlo stochastic simulation was applied to carry out uncertainty statistics of transport time and describe its statistical characteristics.Based on the multi-objective Pareto optimization concept,the Non-dominated Sorting Genetic Algorithm was designed to solve the model.The China-Europe container transport from Tianjin to Hamburg was taken as an example,and the Pareto-optimal solution was obtained by determining the relevant parameters based on the actual research results in order to validate the model and algorithm.The results show that different schemes yield different transport cost,time duration,and carbon emission,and satisfy the definition of Pareto optimal solution.Therefore,the correctness and feasibility of the multi-objective service network design modeling and algorithms considering uncertainty were proved.The results ca

关 键 词：多式联运 服务网络设计 多目标规划 时间不确定性 快速非支配排序遗传算法 

分 类 号：U294[交通运输工程—交通运输规划与管理]