基于分布式混合遗传算法的动态泊位分配策略与仿真

Strategy and simulation on berth allocation based on distributed hybrid genetic algorithm

针对集装箱码头的泊位分配问题,提出滚动式优化决策策略,建立基于整数规划的动态泊位分配模型,以连续岸壁线泊位分配为基础,使模型更加接近实际.同时,在该模型的基础上建立码头前沿的泊位仿真模型,采用分布式遗传算法和启发式算法相结合的分布式混合遗传算法求解泊位分配模型.通过对某集装箱码头的案例分析,说明该泊位分配模型和算法的有效性与实用性.

To the problem of berth allocation, a rolling-horizon approach is proposed. According to this approach, a dynamic allocation model is initially developed based on the integer programming for berth allocation, which is more closed to reality as the result of basing continuum quayside. At the same time, a simulation model is developed based on the berth allocation model, and a distributed hybrid genetic algorithm is employed for solving the model, which combines distributed hybrid genetic algorithm and heuristic algorithm. A case applying this approach on a certain container terminal shows the validity and usefulness of the model and algorithm.