Abstract: The quality of humidity sensitive items is easily affected by the storage environment,rational preparation of location assignment plan for humidity sensitive cargoes in stereoscopic warehouse is an important part of lean warehouse management.The humidity variation coefficient was introduced to construct the humidity calculation function of the storage in the warehouse.An optimization model for storage allocation of humidity sensitive cargoes in warehouses was established to minimize the inlet and outlet distance and maximize shelf stability.The operation strategies of selection,crossover,mutation and immigration were designed,and a comprehensive stereoscopic warehouse storage allocation method for humidity sensitive cargoes based on Multi-population Genetic Algorithm (MPGA) was proposed.The calculation results showed that compared with the traditional genetic algorithm and the single population adaptive genetic algorithm,the proposed method could effectively compile the storage allocation plan of humidity sensitive cargoes stereoscopic warehouse.The optimal solution had low dependence on the value of algorithm parameters,remarkable optimization effect and strong adaptability,which could provide strong decision support for the lean management of humidity sensitive cargoes stereoscopic warehouse.

Key words: location assignment, self-adaptation, multi-population genetic algorithm, humidity sensitive cargoes, comprehensive stereoscopic warehouse

摘要： 湿度敏感类物品的品质易受仓储环境影响,合理编制湿度敏感类物品立体仓库货位分配计划是实现仓库精益化管理的重要内容。引入湿度变化系数,构造立体仓库货位湿度计算函数,以最小化物品出入库距离、最大化货架稳定性为目标,建立湿度敏感类物品立体仓库货位分配优化模型;设计选择、交叉、变异和移民操作策略,提出一种基于自适应多种群遗传算法（MPGA）的湿度敏感类物品立体仓库货位分配方法。计算结果表明:与传统遗传算法、单种群自适应遗传算法相比,所提方法能更有效地编制湿度敏感类物品立体仓库货位分配计划;最优解对算法参数取值依赖程度低,优化效果较显著,自适应性能力强,能够为湿度敏感类物品综合性立体仓库精益化管理提供强有力的决策支持。

关键词: 货位分配, 自适应, 多种群遗传算法, 湿度敏感类物品, 综合性立体仓库