一种基于模糊C均值算法确定制造单元数的方法

A method based on FCM algorithm determining the number of manufacturing cells

单元数的确定是单元构建问题的关键。基于模糊C均值算法,分析已有聚类有效性函数的缺陷,针对单元构建问题的特征,提出新的聚类有效性函数和确定最优单元数的算法。新的聚类有效性函数和算法考虑了单元内的设备数、零件数及成组功效,反映了单元构建的实质,能够排除单元内只有零件没有设备的不可行分组方案。通过多组文献数据测试比较,验证了算法的有效性。

The number of cells is a key factor of Cell Formation（CF） problem. After analysis the defects of currently validity function, a new validity measure and a algorithm which can obtain the optimum number of groups are both proposed, these are based on the Fuzzy C-Means （FCM） algorithm and the feature of CF problem. The new validity measure take into account the number of machines and parts in the cells,the essence of CF is measured by group efficiency. It can eliminate infeasible solution which there is no machine in the cells. The new algorithm is experimentally demonstrated using several well known CF examples from the published literature.