Automotive Industry Revolution on Predictive Analytics and Machine Learning for Vehicle Behaviour and Safety With Environmental Criteria
Anil Kumar Adike, S. Silvia Priscila · 2025
Existing vehicle incident prediction systems in the automotive industry are generally reactive, for instance, based on accident data. This is insufficient because there is no mechanism for proactively predicting where incidents may occur. However, these are not the best solutions to obtain safety objectives because of the limitations of integrating with real-time data sources and scalability. The proposed system uses machine learning (ML) methods such as Random Forest, XGBoost, and Deep Neural Networks (DNN) to anticipate automobile incidents by analyzing real-time data from Internet of Things (IoT) sensors, GPS, traffic cameras, and weather stations. These differ from normal models in that they allow for the entry of dynamic data streams to perform proactive safety measures. The results show significant improvements, such as increased accuracy from 83.3% to 92.1% and reduced false positives and negatives.