Reliability-Informed Life Prediction for New Energy Vehicle Components

For new energy vehicle components, remaining service life requires crucial maintenance, but operating conditions and incomplete fault records still limit the sustainable development of models. This study constructs a data-driven framework for batteries and traction motors by integrating fault analysis, reliability parameter estimation, and machine learning-based prediction. This framework can support maintenance planning and spare parts planning to some extent. However, these results should be interpreted with caution because component type, brand coverage, and data quality may introduce unobserved biases. More extensive cross-brand datasets, prediction ranges that account for uncertainty, and real-time validation are needed before reliable large-scale deployment.