The advent of Industry 4.0 has transformed reliability engineering, especially in predictive maintenance and real-time condition monitoring. This study investigates state-of-the-art methods that utilize data-driven approaches and machine learning algorithms to address modern industrial challenges. This transformation is fueled by the widespread use of interconnected sensors and the rapid expansion of data volumes. Predictive maintenance has evolved towards condition-based strategies, enabling proactive interventions and cost reductions. Real-time condition monitoring benefits from sensor advancements and computational capabilities, allowing for continuous equipment health monitoring.

Another research focus is on redundant systems like MooN (M out of N) and parallel architectures, which are fundamental in safety-critical industries such as aerospace, rail transportation,and autonomous driving, ensuring uninterrupted operation despite component failures. These architectures feature fault tolerance and self-diagnosis capabilities, with fail-operational systems maintaining functionality even when faults occur.

Monte Carlo simulation of series, active parallel and standy-by systems

Reliability of coupled Markovian MooN systems   arxiv:2210.04040

Monte Carlo simulation of MooN system with dependent components

Linear and nonlinear regression model for bi- and multivariate analysis

Predicting lane-changing maneuvers on European highways   j.physa.2023.128471

RUL prediction of simulated aircraft turbofan   Nasa-repository, sample 6