Abstract: The container allocation costs account for more than 20% of the total operational costs for carriers. The rational allocation of containers can greatly improve the transportation efficiency, reduce the transportation costs, and give full play to the advantages of container logistics. Considering the uncertainty of container demand and opportunity loss caused by lack of empty containers or ineffective management, the problem of full and empty container allocations is formulated as a stochastic programming model, which aims at maximizing the gross benefit of the transportation system. The algorithm combines with the dynamic programming (DP) and genetic algorithm (GA) is designed to solve the problem. By numerical experiments, the paper reveals some influencing mechanism of the relative parameters on the allocation strategy.

Key words: integrated transportation, container allocation, dynamic programming, genetic algorithm, stochastic programming

摘要： 集装箱调运成本占承运人运营成本的20%以上.对集装箱进行合理调度，有利于提高运输效率、降低运输成本，充分发挥集装箱物流的优势.本文在考虑重箱运输需求不确定性、航运企业因空箱不足及调度不周所导致机会损失的基础上，建立了多式联运下以运输系统总收益最大为目标函数的集装箱重空箱联合调度随机规划模型，并设计动态规划与遗传算法相结合的组合算法对模型进行求解，实现计划期内的最优动态调度.通过仿真实验分析，揭示了弃货成本、租箱成本、不确定性重箱运输需求分布情况对集装箱调运策略的影响机制.

关键词: 综合交通运输, 集装箱调度, 动态规划, 遗传算法, 随机规划