Enhancing Reliability through AI/ML-Driven Early Failure Mode Detection for Vehicle Engine Mounts

The integration of Artificial Intelligence (AI) and Machine Learning (ML) into automotive engineering is redefining reliability analysis by enabling early failure detection and predictive maintenance. This study presents an AI/ML-based framework for forecasting the operational lifespan of engine Torque rolling restrictor (TRR) mounts critical components responsible for vibration isolation and structural integrity in vehicles. Using supervised learning techniques and terrain-specific operational data, the proposed approach predicts the remaining distance a vehicle can travel before TRR mount failure becomes likely. A Random Forest Regressor model trained on historical usage data achieves high predictive accuracy, with a Mean Squared Error (MSE) of 0.05 and an R² score of 0.85. The results highlight the potential of data-driven techniques to proactively manage component health, reduce downtime, and improve reliability in automotive systems. This case study serves as a blueprint for embedding AI/ML into vehicle design, enabling smarter diagnostics and more resilient vehicle architectures.