Car Engine Health Monitoring System Using Machine Learning

Abstract - This review paper explores the advancements in intelligent engine health monitoring systems that integrate machine learning, explainable artificial intelligence (XAI), and predictive maintenance strategies. The study reviews various approaches for analyzing critical engine parameters such as engine RPM, lubrication oil pressure, fuel pressure, coolant pressure, oil temperature, and coolant temperature to assess engine health and detect anomalies. The paper evaluates the performance of multiple machines learning algorithms, including Random Forest, Gradient Boosting, Support Vector Machines (SVM), K-Nearest Neighbors (KNN), and Logistic Regression, in classifying engine conditions as healthy or faulty. Special emphasis is given to the use of feature engineering techniques, such as temperature difference and pressure ratio, to enhance model accuracy. Furthermore, the review highlights the importance of Explainable AI methods, particularly SHAP (SHapley Additive exPlanations), in providing transparency by identifying the contribution of individual features toward fault prediction. This improves trust and interpretability in AI-driven diagnostic systems. The integration of web-based frameworks like Flask for real-time user interaction and RESTful APIs for mobile application support is also examined. Additionally, the study discusses database-driven systems for storing vehicle-specific parameters and enabling predictive maintenance through automated scheduling and reminders. The paper concludes that combining machine learning with explainable AI and intelligent safety recommendation systems significantly enhances fault detection accuracy, reduces unexpected engine failures, and supports proactive maintenance. These systems have strong potential for real-world applications in automotive diagnostics, fleet management, and smart transportation systems. Key words: Car Engine Monitoring, Machine Learning, Random Forest, Data Preprocessing, Low-pass Filter, Algorithm Comparison Introduction.