基于模拟退火多种群遗传算法的港口船舶调度优化

Vessel Scheduling Optimization Based on Simulated Annealing and Multiple Population Genetic Algorithm

为协调港口航道与泊位资源,提高港口船舶调度效率,从单向航道出发,根据先后调度的2艘船舶的进出港方向和所停靠泊位的远近区分两船间的相对关系,建立以总等待时间最少为目标的调度优化数学模型。设计适用于港口船舶调度优化的模拟退火多种群遗传算法(Simulated Annealing and Multiple Polulation Genetic Algorithm,SAMPGA),模拟某港口不同调度规模的船舶进行仿真试验,与先到先服务规则(First Come First Served,FCFS)和简单遗传算法(Simple Genetic Algorithm,SGA)进行比较,证明SAMPGA在解决航道和泊位协调调度问题上的适用性。结果表明:在现有的调度规则下对航道和泊位进行协调调度能减少船舶的等待时间和总调度时间,但实际调度规则需要考虑的限制因素更多,需对模型作进一步优化。

In order to improve the operation efficiency of the vessels and ports a mathematical model is established reflec- ting the relation between the vessel scheduling and the channel-berth allocation on the assumption of one-way channel. After determining the positional relationship between two successive vessels based on their sailing directions and the distances to the berths, this model solves the problem by means of the Simulated Annealing and Multiple Population Genetic Algorithm （SAMPGA） with the objective of minimum total waiting time. Simulation tests with vessels of different scales are conduc- ted. The results indicate that the SAMPGA is more effective in coordinating channel and berth resources than the Simple Genetic Algorithm （SGA） or the ＂First-come, First-served＂ principle.