The effect of non-normality on performance of linear profile monitoring

Interpretation of quality control charts is usually based on the assumption that successive observations are independent and identically distributed over time. In most statistical process control (SPC) applications, it is assumed that the quality characteristic or vector of quality characteristics of interest follows a univariate or a multivariate normal distribution, respectively. However, in certain cases this assumption may not be realistic. In this paper, we study the effect of non-normality when the quality of a process or product is characterized by a linear profile. Skewed and heavy-tailed symmetric non-normal distributions are used to evaluate the results numerically. The results reveal that non-normality has a significant effect on performance linear profile monitoring when the process is statistically in-control.