This PhD project focuses on developing the next generation of sustainable, self adapting machine learning (ML) systems capable of operating reliably in real world environments that change continuously over time. Building on previous work using Exponentially Weighted Moving Average (EWMA) for covariate shift detection, this PhD will develop an advanced auto adaptive ML framework integrating change point detection (CPD) with modern continual learning strategies. A key feature is the development of real time digital twins. The framework will be applied in brain–computer interfaces (BCIs) and predictive maintenance in smart manufacturing, particularly in photonic device fabrication.

This PhD project focuses on developing the next generation of sustainable, self adapting machine learning (ML) systems capable of operating reliably in real world environments that change continuously over time. Building on previous work using Exponentially Weighted Moving Average (EWMA) for covariate shift detection, this PhD will develop an advanced auto adaptive ML framework integrating change point detection (CPD) with modern continual learning strategies. A key feature is the development of real time digital twins. The framework will be applied in brain–computer interfaces (BCIs) and predictive maintenance in smart manufacturing, particularly in photonic device fabrication.