An Algorithm to Detect and Predict the Remaining Useful Life (RUL), and an Exploratory Data Analysis (EDA) of the Traction Motor Performance

This study investigates and formulates a structured framework to guide deci-sion-making in the remanufacturing of spent electric vehicle (EV) traction motors. With a projected increase in the number of motors reaching end-of-life, the study tack-les key challenges by introducing a data-driven, integrated remanufacturing approach. A predictive algorithm was developed to estimate Remaining Useful Life (RUL) from operational parameters, supported by exploratory data analysis (EDA) that highlight-ed strong inverse relationships between lifespan and stress-related variables such as mechanical load, vibration, and thermal exposure. Various machine learning models including random forest, gradient boosting, and support vector regression yielded moderate predictive performance (mean absolute error ≈ 9.0 km; R² ≈ 0.58), with long short-term memory (LSTM) networks outperforming others (error ≈ 9.1k km; R² ≈ 0.61). These results demonstrate the viability of using predictive analytics to inform reman-ufacturing decisions, contributing to circular economy principles through sustainable EV motor reuse.