The role of transport in sustainable development has been recognized globally. The anchor points for sustainable transport, in the context of infrastructure, public transport systems, good delivery networks, affordability, efficiency, and convenience of transportation, as well as improving urban air quality and health, and greenhouse gas emissions, have been discussed at the global platform—Sustainable Development Summit. This has led to the promotion of Electric Vehicles as a sustainable means of transportation these days. The Indian Government has set a target of 30% electrification of the country’s vehicles by 2030. With advancement in the technology, it’s becoming important to ensure the reliability of such systems. From point of view safety of Electrical Vehicles with an average of 16 EV and hybrid fires per year, there is a 1 in 38000 chance of fire. The reliability and protection of EV systems are of prime importance. The comprehensive study of fault prediction has been carried out in this paper. The Simulink model of EV system is developed for data generation under different defined faults, pattern analysis and prediction of the faults using machine learning algorithms is carried out. Support Vector Machine (SVM), Random Forest, One-vs-Rest (OvR), and Stochastic Gradient Descent (SGD) have been used for the prediction of faults. The result demonstrates the successful application of predictive analysis in problem detection and diagnosis, highlighting its potential to enhance the safety and effectiveness.