Short-run control chart for monitoring Gamma regression profiles

Abstract—In short-run production environments, it is challenging to establish reliable historical data sets for setting control limits and estimating process parameters due to insufficient data. Short-run control charts are used to monitor processes at the start-up stages with no enough initial samples. On the other hands, in many cases, the quality of a process or product is characterized by a functional relationship known as a profile. Typically, the response variable is assumed to follow a normal distribution in research studies. However, in some specific applications, the normality assumption may not hold true, and may follow other exponential family distributions. In these situations, the generalized linear models are used to describe the profile. In this respect, this paper focuses on a Gamma regression profile due to its applications. We develop a self-starting control chart to monitor the model parameters. Then, we appraise the proposed self-starting control chart through simulation studies in terms of average run length criterion under different shifts.

کلیدواژه ها:

Keywords—short-run production ، concurrent monitoring ، Gamma regression profile ، on-line monitoring