ABSTRACT
Due to the quick determination of the root causes for an out-of-control process, the estimation of the starting time of a process disturbance becomes an important issue for the process improvement. In contrast to the typical assumption that the process is normal, this study realizes that the process may follow a gamma distribution. For this reason an [Xbar] control chart was employed to monitor the gamma process. When an out-of-control signal is triggered, we propose a fruitful approach for estimating the starting time of a process disturbance. A series of simulations were conducted to evaluate the performance of the proposed approach.