随着国民经济和交通运输业的快速发展，综合运输体系不断完善，然而运输方式的结构性矛盾依旧较为突出，货运中公路运输的比重过大，运输方式之间衔接不畅，综合运输效率不高。多式联运是助力运输结构调整的有力工具，也是国家近年来的重点发展产业。长大货物具有价值大、风险大等特点，主要运用大型工程及国防建设等项目中，关系国计民生。长大件的运输过程可能会受到桥梁载重、净空高度、路线转弯半径限制等因素影响。结合运输项目自身特点及运输需求，考虑运输过程中的改造工作，从而制定一套合理有效的运输方案，对于长大件货物的物流发展，促进工业经济及国防建设，具有重要的现实意义。本研究通过对国内外现有长大货物运输路径规划，多式联运、多目标优化问题相关文献的梳理与分析，将多目标下的长大货物多式联运路径规划问题作为研究对象，针对现实场景中长大件货物的运输过程中会对沿途设施进行改造的特性，构建了以运输费用最低、运输时间最短、运输过程碳排放量最少和运输方案可靠性最高为目标组合的多式联运路径规划模型。本研究对多式联运运输路网结构进行了重构，包括扩增虚拟节点及生成邻接矛盾矩阵，基于第三代非支配遗传算法对长大货物多式联运费用-时间-碳排放-可靠性模型进行了算法求解设计，建立NSGA-III算法框架，得到问题Pareto最优解集。最后通过算例利用python程序对运输路径的费用、时间、碳排放和运输可靠性目标进行求解，得到了一组Pareto最优解集，以便决策者根据实际需求选择合适的方案，同时将运输过程不考虑改造的问题模型与本研究模型进行方案求解对比，验证了本研究所建模型的合理性与有效性。
With the rapid development of the national economy and transportation industry, the comprehensive transportation system has been continuously improved. However, the structural contradictions of transportation modes are still prominent. There exits the excessive proportion of highway transportation in freight transportation, poor transfer between vehicles, and low overall transportation efficiency. Multi-modal transport is a powerful tool to help adjust the transportation structure, and it is also the national key industry recently. Oversized cargos with the characteristics of high value and high risk are mainly used in large-scale engineering and national defense projects, which is beneficial to national economy and people‘s livelihood. The transportation of oversized cargos may be affected by many factors such as bridge load capacity, net height, and turning radius of the route. Considering the reconstruction of transportation network based on the actual requirements of transportation projects, a reasonable and effective transportation plan is of great practical significance for the logistics development of oversized cargos, promoting industrial economy and national defense construction.This article combs and analyzes the existing domestic and foreign studies including oversized cargo transportation route planning, multi-modal transportation, and multi-objective optimization, takes the oversized cargo multi-objective multi-modal transportation route planning problem as the research object. Considering the reconstruction of transportation network, a multi-modal transport path planning mathematical model that minimizes cost, carbon emissions, and time and maximizes reliability is established. This article reconstructs the multi-modal transport road network structure, including expanding virtual nodes and generating adjacent conflict matrices. Based on the third-generation non-dominated genetic algorithm, a designed algorithm is used to obtain the Pareto optimal solution set of the problem. Finally, a set of Pareto optimal solution sets is obtained by using Python programs to solve the model through numerical examples, which is helpful to the decision-makers to choose appropriate transport plans according to actual needs. Last but not least, this article compares the solutions of the transportation model without considering reconstruction with that of the above model, which is helpful to verify the rationality and effectiveness of the model considering reconstruction.
