Statistical Process Control Based on Two Kinds of Feedback Adjustment for Autocorrelated Process

As manufacturing quality has become a decisive factor in competing in a global market, statistical process control (SPC) is becoming very popular in industries. With advances in sensing and data capture technology, large volumes of data are being routinely collected in automatic controlled processes. There is a growing need for SPC monitoring and diagnosis in these environments, SPC (statistical process control) and APC (automatic process control) can be integrated to produce an efficient tool for process variation reduction. In this paper, we discuss the monitoring of MMSE (minimum-mean-squared-error-) and PI (proportional-integral-) controlled processes. Then control charts performances of process output and control action of two kinds controlled processes are compared by ARL (average run length). A simple example is used to illustrate monitoring methods. We show that which chart should be used for different feedback adjusted processes.