约束规划求解自动化集装箱码头轨道吊调度

Constraint Programming for Scheduling Rail-Mounted Gantry Crane in Automatic Container Terminal

针对在自动化集装箱码头(Automated Container Terminal,ACT)堆场中单箱区内配置双自动化轨道式龙门起重机(Automatic Rail-Mounted Gantry Crane,ARMG)的调度问题进行研究,为ACT找到合理、高效的集装箱搬运操作顺序。引入作业干扰、任务接力和接力箱区容量等约束,建立以最小化集装箱搬运总延误时间为目标的约束规划(Constraint Programming,CP)模型。利用CP建模的灵活性,将3种调度规则设置在求解过程中,对港口多种任务规模数量进行算例分析。结果表明:在小中规模任务数量情况下,海侧ARMG任务优先规则能在10 s内得出求解结果;在大规模任务数量情况下,接力箱任务优先规则在求解质量上至少提升30.87%,求解时间减少50%。

The container handling efficiency of an ACT(Automated Container Terminal)is improved through solving the CP(Constraint Programming)problem of ARMG(Automatic Rail-Mounted Gantry Crane)scheduling for the single-container-block area of the terminal yard.The constraints associated with the scheduling problem,such as disturbance to operation,task relaying and the capacity of the site for task relaying,are introduced and the CP model for minimizing the total container handling time is established.The model is solved for cases of different throughput with three scheduling rules respectively.It is found that under the condition of a small and medium-scale task,the problem can be solved within 10 s with the sea-side ARMG task priority rule,while in the case of a large-scale task,using the relayed box task priority rule can improve the solution quality by at least 30.87%and can reduce the computing time by 50%.