This study presents an integrated approach to vehicle vibration pattern analysis and early fault detection using Artificial Neural Network (ANN) modeling to address key maintenance challenges, including unplanned engine downtime and the limitations of conventional time-based preventive maintenance strategies. The research combines empirical vibration measurements obtained from internal combustion engines with a computational ANN model developed to automatically classify engine health conditions. The experimental phase focused on collecting vibration data under three operational states: normal, degrading, and faulty, each exhibiting distinctive vibration characteristics associated with combustion pressure variations and mechanical imbalance.A total of 500 vibration samples were recorded, with 400 used for ANN training and 100 for validation. The proposed feedforward multilayer perceptron (MLP) model achieved a classification accuracy of 94% with a mean squared error of 1.5 × 10⁻⁵. The ANN successfully identified transitional degradation patterns between normal and faulty states, demonstrating its capability for early fault diagnosis and supporting timely maintenance intervention and improved maintenance scheduling. Furthermore, the results reveal a strong correlation between combustion chamber pressure fluctuations and vibration amplitude, confirming that increased pressure irregularities lead to higher vibration energy levels. Overall, the proposed ANN-based vibration monitoring framework provides a non-invasive, cost-effective, and reliable solution for real-time engine condition assessment. By enabling early detection of mechanical degradation and reducing the risk of unexpected failures, the approach contributes to enhanced equipment availability, lower maintenance costs, and the effective implementation of predictive maintenance systems in automotive and industrial applications.