A Review on Predictive Fault Detection in Pre-Release Devices and Systems

Modern system development across sectors like automotive, aerospace, and electronics has become highly complex, leading to increased risks of design-stage faults that traditional testing fails to detect early. These reactive methods result in costly recalls, delays, and compromised safety. This paper introduces a proactive solution for predictive fault detection in pre-release systems using artificial intelligence, digital twins, and real-time monitoring. The proposed system simulates system behavior in the design phase, identifies potential failure points using trained machine learning models, and validates predictions through IoT sensor data. By addressing faults before physical prototyping, the method improves product reliability, reduces development costs, and shortens time-to-market. The paper outlines the methodology and potential of this approach to be scalable across various industries requiring high-reliability systems.