The exponentially weighted moving average (EWMA) control charts are widely used to detect abnormalities or shifts in a given process. Meanwhile, some processes require great attention at the initial setup to ascertain that the process is free from any outliers or disturbances. It is shown in this article how such a process can be monitored or handled so that any outliers or abnormalities in the process can quickly be detected, and corrective action can thereafter be taken. The newly proposed method is highly sensitive to detecting small and moderate outliers or disturbances. The newly proposed chart can be used by quality and reliability practitioners and other professionals to monitor processes that require careful attention at the initial setup. This will prevent material loss that can occur as a result of indifference to earlier process changes or disturbances. The newly proposed method is compared to other existing methods using average run length, that is, the average time required to detect the outliers or shifts. The performance comparison shows that the newly proposed method requires a shorter time to detect any disturbance or outliers or shifts in a process at the startup. The application of the proposed method to a real-life situation is also presented.
The article featured today is from Quality and Reliability Engineering International with the full article now available to read here.
Generalization of time-varying fast initial response for exponentially weighted moving average control charts. Qual Reliab Eng Int. 2022; 1– 12. https://doi.org/10.1002/qre.3194, , , , , .