Diagnosis of Out-of-Control Signals in Multivariate Manufacturing Processes with Random Forests
In a multivariate manufacturing process, there are two or more correlated quality characteristics which are needed to monitor and control simultaneously and hence various multivariate control charts are applied to determine whether a process is in-control. Once a control chart gives an alarm, the next work is to determine source(s) of the out-of-control signals. In this paper, a random forests model is developed to diagnose source(s) of out-of-control signals in multivariate processes. A bivariate manufacture process is conducted to compare the performance of the random forests model with a support vector machine (SVM) model. Results show that the random forests model is better than the SVM model. The results also indicate that the effectiveness of the random forests model in identifying the source of out-of-control signals.
KeywordsManufacturing quality control and management Predictive maintenance Diagnosis and prognosis of machines Random forests