Abstract: Based on the synchronized zoning strategy in automated zone-picking system，an item assignment problem is proposed to balance workload between zones and improve the picking efficiency.It is classified as a clustering problem in which the number of clusters is predefined.In order to solve this problem，a dynamic clustering algorithm applied in manual picking system is used as a source of reference.Based on the features of automated zone-picking system，the algorithm is improved in two aspects.The first aspect is that the influence of amount of goods on similarity coefficient is considered.The other aspect is that the Max-Min Ant System（MMAS） is combined with the clustering algorithm.At last，a simulation analysis is made on the two clustering algorithms，and the results prove that the system has a higher efficiency improvement in the optimized clustering algorithm.

Key words: automated picking system, synchronized zoning, item assignment, similarity coefficient

摘要： 基于自动拣选系统中并行拣选策略，提出货物品项分配优化问题，以平衡各拣货区工作量，提高系统拣选效率。该问题可归结为分类数已知的聚类问题。为解决此问题，借鉴了人工拣选系统中解决该问题的动态聚类方法，并针对自动拣选系统的特点做两方面改进：在品项间相似系数表达中考虑了各品项拣选数量的影响；在聚类过程中结合了最大最小蚁群算法（MAX-MIN Ant System，MMAS）。最后对两种聚类方法进行仿真分析，结果证明了改进的聚类方法可更大幅度地提高系统拣选效率。

关键词: 自动拣选系统, 并行拣选, 品项分配, 相似系数