Track: Quality Control and Management

Abstract

Quality control and improvement have become one of important aspects in assuring production sustainability. Quality has been considered as one of  competitive advantages for a company. One of main techniques in conducting quality control is control chart. Early development of this control chart was univariate. However, product quality can not be justified only by one product characteristic. Hence, multivariate statistical process control (MSPC) is necessary to be applied in order to monitor many product quality variables simultaneously. It is important to explore more accurate methods for MSPC modelling. Moreover, rapid development in process automation and computer technology has shifted quality control process, from product quality monitoring into process monitoring. Real time process monitoring for many process characteristics has been conducted to ensure optimal process. Automatic data acquisition has provided massive data stream : large and complex data. These phenomena have supported the application  of data mining techniques for process monitoring. This paper focuses on reviewing previous literature  related to MSPC, from traditional to data mining based methods.  This study aims to synthesize previous literature in multivariate control charts for process monitoring, hence the area for future research is identified