Enhancing process stability with robust charts using high breakdown point estimators in contaminated settings: an application in healthcare monitoring*

Control charts are essential in medical processes as they help monitor process stability by detecting variations over time, allowing for timely interventions to maintain and improve health quality. Commonly, it is assumed that the unknown parameters are estimated from Phase I clean data. There are numerous practical scenarios where this is not always true. In this study, we propose different control charting schemes by using the robust dispersion estimators of high breakdown point in Phase I to handle this issue, especially focusing on diffused and localized variance disturbances. The performance of the developed schemes is evaluated in terms of average run length (ARL) and conditional expected delay (CED) under clean and different normal environments with contaminations applying Monte Carlo simulations techniques. The results revealed that the schemes constructed based on robust estimators perform superior to their existing competitors under contaminated set ups. A real-life application in health care monitoring and two other applications are also investigated to highlight the significance of the study for practitioners’ considerations.