Monitoring Quality Characteristics Using Linear Profile

Control charts are used for monitoring single or multivariate quality characteristics. However, some processes can be characterized by a linear profile. Monitoring of such profiles using traditional control charts may not result in optimum control of the process. This paper provides a brief discussion on the linear profiles and presents a solution based on a multivariate cumulative sum control (MCUSUM) chart for monitoring mean vector of the slope and Y-intercept in combination with R-chart for monitoring the error variance. The performance of the proposed scheme is evaluated using average run length criterion. Simulation results indicate that the proposed scheme performs relatively better than the existing methods.

Keywords:

Linear Profile , Multivariate Cumulative Control Chart (MCUSUM) , Average Run Length , Exponentially Weighted Moving Average Control Chart (EWMA) , Statistical Process Control (SPC)