Lithium-ion batteries are integral to various systems, including electric vehicles and warehouses, where estimating their operational lifespan is critical. This lifespan, known as Remaining Useful Life (RUL), helps in optimizing performance and maintenance. Accurate RUL prediction helps avoid unexpected breakdowns and lowers maintenance costs, making these systems more reliable. This paper introduces a novel approach using a Rotary Positional Embedding (RoPE) Enhanced Informer model to address challenges in long-term time series forecasting for RUL prediction. By integrating RoPE into the Informer, the model efficiently preserves positional information and leverages sparse attention mechanisms to reduce computational complexity, enabling it to handle longer sequences with minimal memory usage. The proposed model was tested on benchmark datasets, where it outperformed peer level approaches in terms of Mean Absolute Error and Root Mean Squared Error. The results show a significant reduction in validation errors, indicating the model’s superior generalization ability and its robustness against overfitting. This approach offers a scalable solution suitable for real-time predictive maintenance, particularly in resource-constrained environments. The combination of advanced positional encoding with the Informer architecture demonstrates the potential for this method to enhance battery management and improve the accuracy of RUL prediction across various industrial applications.