Prediction of Automotive Vehicles Engine Health Using MLP and LR

Engine Health Prediction determines the health of automotive engines. By analyzing various engine parameters, the model aims to identify potential issues and predict engine failures before they occur. This paper introduces an advanced method for engine health prediction in automotive systems using machine learning, specifically Multi-Layer Perceptron (MLP) and Logistic Regression models. The approach focuses on early detection of engine failures and optimizing maintenance schedules, leveraging a dataset with key engine performance indicators. The methodology involves data preprocessing, feature engineering, model training, and validation. The MLP model addresses non-linear data through densely connected layers and regularization, while Logistic Regression offers a probabilistic approach for binary classification. Both models are evaluated on accuracy, precision, recall, and F1-score. The results show that the MLP model is superior in identifying complex patterns, leading to strong predictions. On the other hand, logistic regression provides quick and interpretable results. The paper concludes with an optimized integrated method combining both models’ strengths to enhance prediction reliability. This research not only contributes to the field of predictive maintenance but also offers a practical tool for automated engine health assessment, potentially reducing downtime and maintenance costs. Future research will focus on integrating these models into real-time monitoring systems and adapting them to various engine types through transfer learning.