基于层次聚类的机加工线平衡优化

Transfer Line Balance Optimization Based on Hierarchical Clustering

针对大规模机加工线平衡问题在求解过程中存在算法复杂、效率较低等问题,提出一种针对缸体类零件加工特征的层次聚类方法。为有效减少加工元种类,对缸体类复杂零件特征进行转化,以加工元为最小单位,依据加工元间的相似度值,应用层次聚类方法进行聚类分组。以节拍、平衡率为优化目标,构建基于层次聚类的线平衡问题求解模型,并利用改进的遗传算法进行优化求解,交叉变异算子采用自适应策略,避免遗传算法陷入局部最优。最后以某企业发动机缸体为研究对象验证了方法的有效性。

In order to solve the problem of complex algorithm and low efficiency in the solving process of the balancing in large-scale machining lines, a hierarchical clustering method was proposed to deal with the machining features of cylinder-type parts. In order to effectively reduce the category of processing elements, the characteristics of complex cylinder-type parts should be transformed. The processing unit should be taken as the minimum unit, and the cluster was grouped by the hierarchical clustering method according to the similarity values between processing elements. Taking cycle time and balancing rate as optimization objectives, a model for solving of line balance was constructed based on hierarchical clustering and it was optimized by using improved genetic algorithm for solving. The crossover mutation operator should adopt the adaptive strategy to avoid the genetic algorithm from getting into local optima. Finally, an engine cylinder block of an enterprise was used to verify the validity of the method.