Advancements in Carpooling System using Machine Learning

Carpool programs help reduce emissions and traffic jams offering a solution to transportation challenges. This study explores the effectiveness of using fare prediction models to enhance carpooling services. By combining the Haversine formula with machine learning techniques in our approach we can accurately calculate distances between pick up and drop off points. Our aim is to optimize fare estimates by integrating these predictions into the carpooling system ensuring fairness and transparency for both drivers and passengers. The results indicate that our model significantly boosts the accuracy of estimates compared to methods leading to improved efficiency and customer satisfaction in the carpooling service. This research sets a foundation for developments in intelligent transportation systems, within smart cities.