Abstract :

The rapid growth of app-based bike taxi services has increased the need for accurate demand forecasting to improve service efficiency and customer satisfaction. This project focuses on predicting Ola bike ride request demand using machine learning techniques. Historical ride data is analyzed to capture temporal, spatial, and environmental patterns influencing demand. Features such as time, day, location, weather, and past demand trends are used to train predictive models. Machine learning algorithms enable accurate short-term demand forecasting. The proposed system helps optimize driver allocation and reduce passenger waiting time. Experimental results show improved prediction accuracy compared to traditional statistical methods.