Joint process monitoring in power plants using conditional aggregate mean control charts

Industrial processes are subject to inherent variations that require monitoring to distinguish between natural and unnatural fluctuations. Control charts are the most common tools used in statistical process control to monitor these changes. Traditionally, separate charts are used for monitoring process mean and variance. Recent studies favor joint charts that monitor both parameters, because they improve detection and diagnostic accuracy. We propose a new way to construct control charts based on the Conditional Aggregate Mean (CAM) approach for joint monitoring of process mean and variance. The method integrates three variance transformations, including first application of Johnson four‑parameter transformation within joint memory chart. We show that the proposed charts detect medium shifts faster than existing schemes. A case study on a power plant confirms their reliability and early detection of shifts. The charts suit power plant monitoring as they handle moderate and frequent parameter changes effectively.