带有固定运费的多式联运方式选择  

Solving the Mode Selection Problem with Fixed Transportation Cost in Intermodal Transportation

作　　者：李愈[1,2] 赵军[2] 吴刚[1] 陈佳琪[1] 

机构地区：[1]西南交通大学峨眉校区交通运输系,四川峨眉山614202 [2]西南交通大学交通运输与物流学院,四川成都610031

出　　处：《西南交通大学学报》2012年第5期881-887,共7页Journal of Southwest Jiaotong University

基　　金：教育部人文社会科学研究规划基金资助项目(11YJA790164/10YJA790020);中央高校基本科研业务费专项资金资助项目(10501B10096001)

摘　　要：为更经济有效地组织多式联运,在同时考虑固定运输费用和变动运输费用的前提下,针对一批货物在其运输路径上各路段运输方式的优化选择问题,以包括运输费用、中转费用和惩罚费用在内的总费用最少为优化目标,以货物送达时间和运输能力为约束条件,构建了混合整数线性规划模型,并设计了相应的整数编码遗传算法.算例结果表明,本文模型比既有不考虑固定运费的模型具备找到更优解的能力;对于15个包括20～100个路段和5～20种运输方式的随机算例,与商业优化求解器LINGO相比,本文算法在设定计算时间60 s内找到了2个大规模算例的更优解,总费用平均节省0.75%,证明了该算法对大规模问题的有效性.To organize the intermodal transportation in a more cost-effective way,the problem of selecting the transportation mode in each link along the transportation route of an intermodal shipment was addressed by taking into account the fixed and variable transportation cost simultaneously.Aiming at minimize the total cost including transportation cost,transfer cost,and penalty cost,and taking the shipment delivery time and transportation capacity as constraints,a mixed integer linear programming model was proposed,and a corresponding integer-based encoding genetic algorithm was designed.Computational results show that the proposed model can find much better solutions when compared with the existing model that does not consider the fixed transportation cost.For 15 random instances with 20 to 100 links and 5 to 20 transportation modes,the present algorithm find much better solutions than the commercial optimization solver LINGO within the same computation time of 60 seconds for 2 large-scale instances,achieving an average total cost-saving of 0.75%.This proves the efficiency of the proposed algorithm for large-scale instances.

关 键 词：多式联运 方式选择 混合整数线性规划模型 整数编码遗传算法 固定运费 

分 类 号：U294[交通运输工程—交通运输规划与管理]