The critical importance of rotating machinery and its reliance on rolling element bearings across industrial applications necessitates advanced diagnostic systems to preempt catastrophic failures. Traditional single-fault diagnostic approaches, while useful, fall short in complex industrial settings where multi-fault scenarios are common. This study introduces an Intelligent Multi-Fault Diagnostic System (IMFDS) for shaft-bearing systems, leveraging Artificial Intelligence (AI), the Internet of Things (IoT), and Cloud Computing for real-time, scalable fault detection and classification. The proposed system integrates vibration-based condition monitoring, advanced wavelet transformations, and machine learning algorithms (specifically, Artificial Neural Networks and Support Vector Machines) to detect and classify bearing faults with high accuracy. Experimental validation confirms the diagnostic model’s precision, with SVM achieving high level accuracy over ANN’s, highlighting SVM’s superior generalization in complex fault environments. Additionally, the architecture supports Industry 4.0 initiatives by facilitating IoT-enabled data acquisition, real-time analysis, and cloud-based storage, enabling predictive maintenance. A case study on a reciprocating compressor demonstrated IMFDS’s capacity to monitor diverse fault conditions, validating its accuracy through physical inspection. This research advances fault diagnostics by enabling early intervention, reducing downtime, and extending equipment life cycles, making it a pivotal contribution to modern industrial maintenance and operational efficiency.