Intelligent Prediction System for IoV Automotive Vehicles in Smart Cities Integrating IoT, Cloud Computing, Digital Twin, and Machine Learning

The growth of smart cities demands smarter and more efficient transportation, safety, performance, maintenance, and cost optimization systems to tackle challenges and improve decision-making. This paper introduces an intelligent prediction and simulation solution for the Internet of Vehicles (IoV) designed specifically for automotive applications in smart cities. Combining IoT, Cloud Computing, Digital Twin technology, and Machine Learning, the system gathers data from connected vehicles, processes it in the cloud, and creates accurate prediction models of physical systems. Machine Learning is then used to predict patterns to optimize vehicle engine performance, estimate repair costs, evaluate insurance claims for fraud, and support decision-making in complex prediction scenarios. This approach enhances vehicle lifecycle management and performance, improves safety, reduces costs detects claims frauds, and makes driving more energy efficient. The proposed system offers a comprehensive solution to modern mobility challenges, connecting vehicles with the infrastructure landscape, and providing smarter and more sustainable transportation for maintenance, safety, secured insurance, and performance in the cities of tomorrow. The final output results have an accuracy of about 98% for fraud detection, 96.7% for Vehicle health prediction, and a Mean Square Error of 0.003 for repair cost prediction. This system has proved consistent during all phases, providing valuable and effective support to the smart cities domain.