摘要
针对船舶舾装件立体仓库入库货位分配优化问题,本文建立了以货架稳定原则、齐配就近原则以及出库速度最快原则为优化目标的多目标优化模型。实现了考虑船舶出库不确定特点且兼顾托盘齐配率提升的立库入库优化。该研究通过分布式鲁棒优化解决了由于舾装件种类多、标准化程度低等特点导致的出库频率分布难以通过历史数据进行准确刻画的问题。基于模型的数学结构对模型进行分解,并研究设计了使用遗传算法进行热启动的分支定价算法。克服了启发式算法无法获得精确解、传统优化算法求解时间过长的两大难题,实现了大规模问题的快速精确求解,并通过数值实验结果表明:该模型对舾装件入库货位分配具有明显的优化效果。
Aiming at the warehousing optimization of the storage allocation of ship outfitting,a multi-objective optimization model is established with the principles of shelf stability,goods of the same category placed nearby,and the fastest delivery speed as the optimization objective,the optimization of three-dimensional warehousing considering the retrieval uncertainty of the ship and the improvement in the pallet ready rate.Through distributional robust optimization,this research solves the issue of the difficulty in accurately describing the distribution of retrieval frequency through historical data due to the characteristics of various types and a low degree of standardization.A model-based mathematical structure is used to decompose the model,and a branch pricing algorithm using a genetic algorithm for a warm start is studied and designed.The two major difficulties,namely,the difficulty in obtaining the exact solutions through the heuristic algorithm and the exceedingly long duration needed by the traditional optimization algorithms to solve the model,are successfully overcome.The model provides rapid and accurate solutions to large-scale problems.The numerical experimental results show that the model has an obvious optimization effect on the allocation of the storage location of the ship outfittings.
作者
赵天毅
李娜
李明昊
蒋祖华
ZHAO Tianyi;LI Na;LI Minghao;JIANG Zuhua(School of Mechanical&Power Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;China Merchants Heavy Industry(Jiangsu)Co.,Ltd.,Nantong 226113,China)
出处
《哈尔滨工程大学学报》
EI
CAS
CSCD
北大核心
2024年第1期1-8,共8页
Journal of Harbin Engineering University
基金
工信部高技术船舶项目([2019]360号).
关键词
舾装件
立体仓库
货位分配
不确定性
多目标优化
分布式鲁棒优化
遗传算法热启动
分支定价算法
ship outfitting
three-dimensional warehouse
storage allocation
uncertainty
multi-objective optimization
distributional robust optimization
genetic algorithm warm start
branch pricing algorithm
