A Privacy-Preserving and Biometric-Aware Tasks Reallocation Strategy in Industry 5.0

Industry 5.0 represents an emerging industrial paradigm that emphasizes seamless collaboration between human workers, collaborative robots (cobots), and smart objects. Its goal is to enable intelligent, adaptive manufacturing environments that not only boost operational efficiency and ensure regulatory compliance but also enhance workplace safety. In this context, we designed a complete framework based on a Reinforcement Learning (RL) strategy for intelligent and privacy-preserving task reallocation. Central to our vision is the prioritization of human well-being ensuring that both worker safety and privacy are protected, while the performance and reliability of machines and devices are optimized to support a truly human-centric manufacturing system. Our solution monitors workers' physiological states and detects signs of fatigue, stress, or overload, ensuring that tasks can be dynamically reallocated to another worker or cobot to promote well-being without manual intervention. Moreover, by ensuring biometric data remains local to the worker's device, the system respects data sovereignty and avoids unnecessary sharing of sensitive health information, guaranteeing compliance with regulations like GDPR. Our solution can adapt dynamically to the changing conditions and needs of human operators creating a privacy-preserving, safe, and efficient collaborative environment between people and machines. A comprehensive experimental analysis assesses the accuracy and performance of the proposed approach.