The manufacturing industry is heavily dependent on human labor and ensuring the health and safety of workers is of primary importance. The emergence of digital twin technology has opened up new avenues for creating virtual replicas of real systems. This thesis proposes to develop a digital humanization framework for evaluating health and safety in production processes/systems, which uses advanced sensors, machine learning algorithms, and simulation techniques to create a real-time virtual representation of a worker’s physical state and behavior. The framework aims to provide an objective and accurate assessment of health and safety risks for workers, enabling early detection and prevention of potential hazards.