Designing an Artificial Neural Network for Simultaneous Detecting, Diagnosing and Quantifying the Magnitude of Mean Shift(s) in Multivariate-attribute Processes

In some statistical process control applications، the quality of a product is characterized by the combination of both correlated variable and attributes quality characteristics. To the best of our knowledge، there is no method in the literature available for identifying the shift magnitude in the out-of-control quality characteristics in multivariate-attribute processes. In this paper، a neural network (NN) -based method is proposed to identify the magnitude of shifts in the out-of-control quality characteristics. The proposed methodology can also determine the process state and diagnose the quality characteristic (s) responsible for out-of-control signals. The performance of the proposed NN-based method in determining the mean shifts magnitude، detecting the process change as well as diagnosing the out-of-control quality characteristic (s) is evaluated based on a numerical example through simulation studies. In addition، the performance of the proposed NN in detection and diagnosis is compared with existing methods in the literature. The results of simulation study show the satisfactory performance of the proposed NN