翻箱是影响集装箱堆场作业效率的瓶颈,也是堆场作业调度的老大难问题。在集装箱堆场现行堆存位置和提箱装船顺序既知的条件下,基于集装箱装船过程中堆场翻箱作业的多阶段性特征,将翻箱作业调度优化问题归结为一个动态最短路径问题,据此构建了优化模型.基于并行处理的思想,将顺序递推算法与逆序递推算法相结合进行优化求解,通过嵌入启发式规则选择落箱位置来减少动态规划的状态空间,从而达到减少二次翻箱,降低总翻箱次数的目标,实现翻箱作业调度的优化.仿真算例分析显示了本文提出的算法的有效性与实用性,可为翻箱作业调度优化提供决策支持.
The container relocation is becoming a bottleneck which significantly restricts the operation efficiency of the whole container yard. let is also a difficult problem facing the container yard scheduling. On condition that both the current stock positions and the picking sequence of containers are given during the loading process, the container relocation scheduling is described as a dynamic shortest path problem, based upon the characteristics of the multiple phases of container relocating operations. According to the concept of parallel processing, the sequential recurrence algorithm is integrated with the inverted recurrence algorithm to solve the problem. The state space of dynamic planning is minimized by the embedded heuristic rules to select the optimal positions of the relocating containers. The optimization of container relocation scheduling is realized through avoidance of the secondary relocation for the same container and the minimization of total relocating times. The simulated numerical experiments demonstrate the effectiveness and practicality of the proposed algorithm, which provides decision support for container relocation seheduling optimization.