Embedding a back propagation network into fuzzy c-means for estimating job cycle time: wafer fabrication as an example

摘要: Cycle time is the required for a job to be processed by factory including processing, transportation, and waiting. Estimating cycle of each critical concern in managing factory. To address this concern, classifying approaches which jobs are classified before or after forecasting their times have recently been proposed. However, none these can guarantee compatibility classifier with mechanism. overcome difficulty, new approach proposed study. In methodology, training mechanism embedded into iterations classifier. Consequently, classification stages interweave other, improving compatibility. The methodology was tested using data on 120 jobs. According results, surpassed five existing methods accuracy. Compared two approaches, statistically significantly reduced mean absolute percentage error 56 38 % testing unlearned data, respectively.