SECURE PREDICTIVE MAINTENANCE FOR INDUSTRIAL SYSTEMS USING FEDERATED LEARNING

In this paper, a secure and scalable predictive maintenance approach for Industry 4.0 using Federated Learning (FL) and Artifficial Intelligence (AI) is addressed.Unlike other approaches, FL maintains data on the premises, which guarantees privacy and regulatory conformance.The system design relies on edge devices to train protected local models and share secure updates.It consists of data pre-processing, model training, and secure aggregation.Experimental results demonstrate FL can obtain high efficiency, communication reduction and improve security against cyber threats.System challenges and future directions are also described in the paper.This paper provides a privacy-aware, up-to-date analysis of predictive maintenance in smart industrial environments.