基于综合满意度的多式联运路径模型及其算法

A Model of Multi-modal Transport Path Based on Hybrid Algorithm Considering Comprehensive Satisfaction

针对多式联运各参与者需求偏好不同的实际背景,以提高所有参与者的综合满意度为优化目标,对中欧集装箱多式联运的路径选择问题进行研究。考虑组成列车等待时间的要素,以列车、船舶的固定时刻表和收货人的软时间窗要求,组成的混合时间窗为约束条件,在分析了影响各参与者满意度因素的基础上建立了具有效用值偏好信息的综合满意度模型。为解决货物时、空、量的衔接组合问题,设计双信息素蚁群算法,用以搜索路径与运输方式的组合搭配,并将改进的小生境遗传算法嵌套进蚁群算法。采用由连云港到马德里的实例做对比验算,分别给出不同需求偏好下的运输方案,并与现有研究方案对比,其综合满意度指标可平均提升约6%~10%,采用混合算法后收敛速度平均提升约36.21%,能够为不同参与者在实际运输过程中路径与运输方式的选择提供决策支持。

Path selection for China-Europe multi-modal transport of containers is investigated in the context of actual differences in demand preferences of various participants in multi-modal transport,with an objective of optimizing the comprehensive satisfaction of all participants. Considering fixed timetable of trains and ships,and taking soft time window requirements of consignees as constraints,a comprehensive satisfaction model of utility value preference information is developed based on an analysis of main factors that affect satisfaction of each participant. To solve a problem of cohesive combination of time,space,and volume of cargo,a dual pheromone colony algorithm is designed to search for combinations of paths and corresponding transport modes. To improve the efficiency of the algorithm,an improved niche genetic algorithm is embedded in it. An actual case study of transport path from Lianyungang to Madrid is used for comparison and verification. The transport schemes under different demand preferences,and compare with the existing schemes,the comprehensive satisfaction index can be increased by 6% to10% on average,and the convergence speed after using hybrid algorithm can be increased by 36.21% on average.It can provide decision support for different participants to choose the optimal path and modes of transport in actual transport process.