Abstract: To solve the low transport efficiency of International containers caused  by small production,variety and unpredictable demand,aiming at the cooperative transportation of International containers with multi-code cargo handling,according to the difference between actual demand and estimated demand,a trust based collaboration model for relative routes with two types of participations who were trust partners and semi-trust partners was constructed,and a production rule based genetic algorithm to match transport tasks and service capabilities with constraints of ship capacity and routing set from angle of shipping service capability was proposed.The main point of the algorithm was segmenting the demand of each pair of ports into small pieces and then allocating the recombined tasks to members in collaboration.By conducting the simulating study on Mid-East cargo shipping services,the results demonstrated that the proposed model was efficient and promising,such as the maximum overall profit of collaboration was largely increased.

Key words: shipping service, trust, collaboration model, genetic algorithms, production rule

摘要： 为解决批量小、平均生命周期短、种类多、需求不可预测引起的集散型国际货运服务的运输效率低的问题,针对多节点可装卸货物的国际集装箱协同运输建立了混合信任模型,以优化协同网络的运输效率。首先依据协同成员对联盟的不同信任程度,将船运联盟中的船运公司分为信任型和半信任型两类协同参与成员,构建了相关航线的混合信任协同模型,提出按协同参与成员信任属性汇总运输需求和运输能力,然后按“单位需求”分解运输需求,在航线、船容量等约束条件下,运用融合产生式规则的遗传算法对运输任务与服务能力进行动态匹配,求解服务协同网络总体利润最高的匹配方案。以义乌小商品出口海运中的中东航线实际运输问题为例,对模型进行验证、求解,并比较分析了多组试验结果,总结讨论了模型的适用范围及其局限,并提出了模型的改进方向。

关键词: 船运服务, 信任, 协同模型, 遗传算法, 产生式规则