Estimating Battery Health of 72V/20Ah Lithium-Ion Batteries in Indonesian Electric Motorcycles Using a Decision Tree Model

The growing popularity of electric motorcycles in Indonesia emphasizes how important it is to monitor battery health effectively in order to guarantee performance, safety, and dependability. This study employs the Decision Tree algorithm, a machine learning technique renowned for its ease of use and interpretability, to forecast the State of Health (SoH) of 72V/20Ah lithium-ion batteries. The model was trained and validated using key battery metrics, including voltage, current, temperature, and capacity. The results show that the model achieves high accuracy, with a Mean Squared Error (MSE) of 4.2961 (0.1072%), a Mean Absolute Error (MAE) of 1.5475 (2.4526%), and a coefficient of determination (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">${R}^{2}$</tex>) of 0.9954. These findings indicate that the Decision Tree approach is effective for real-time battery health estimation and can be integrated into Battery Management Systems (BMS). Although the model performs well, it may be sensitive to overfitting, especially under changing battery conditions. This suggests the need for further refinement or use of ensemble methods. Overall, this research provides a practical and lightweight solution for improving battery monitoring in electric motorcycles and supports the broader adoption of electric vehicles in developing regions.