On Distribution and Average Run Length of a Two-Stage Control Process

In this article a method, using finite Markov chains, to obtain the run-length properties of a two-stage control process is presented. The method furnishes the obtaining of the distribution of waiting time to signal that gives additional insight into the design and performance of a control chart when a warning zone is considered to feature a two-stage control process and when a departure from the null assumption can be clearly defined. An example is given for illustration when samples come from a normal population, though not necessary, with an outlined process inspection scheme. A second example is given to demonstrate the extension of our approach to modelling Markov dependent data observations.