MACHINE LEARNING BASED-BUS ARRIVAL TIME PREDICTION USING HISTORICAL DATA

In this paper, we designed a machine learning-based system for bus arrival time prediction. It combines the use of historical and real time traffic data. The developed approach effectively addressed the drawbacks of static timetables or just real-time GPS tracking by taking various influential contextual parameters, for example time of the day, level of congestion or route, into account. Appropriate data pre-processing and feature engineering enabled our machine learning models to recognize intricate patterns from the traffic data. Several machine learning models including Linear Regression, Decision Tree, Random Forest, Support Vector Regression and Gradient Boosting were investigated and compared. Gradient Boosting model achieved the highest prediction accuracy with lowest error among all models which proved to be a great candidate for this task. This comparison demonstrates the superiority of ensemble learning models in dynamic and complex traffic situations. A simple interface was also designed, facilitating bus passengers and traffic managers to access predicted bus arrival time dynamically. This work provides a foundation for improvement of bus arrival time prediction system through machine learning for enhancing efficiency of public transport.

[1] D. Sun et al., “Predicting Bus Arrival Time on the Basis of GPS Data,” 2007.[Link] [2] R. Jeong and L. R. Rilett, “Prediction Model of Bus Arrival Time for Real-Time Applications,” 2005. [Link] [3] W. H. Lin and J. Zeng, “Experimental Study of Real-Time Bus Arrival Pre-diction with GPS Data,” 1999. [Link] [4] C. Lee and Y. Yoon, “Bus Arrival Time Prediction using Deep Learning,”2022. [Link] [5] A. Agafonov and A. Yumaganov, “Bus Arrival Time Prediction using LSTM,”2019. [Link] [6] “Bus Arrival Time Prediction based on Network Model,” 2017. [Link] [7] “Bus Arrival Prediction using Network Data Model and Time Periods,” 2020.[Link] [8] P. Pov et al., “Bus Arrival Time Prediction Using Machine Learning Ap-proaches,” 2025. [Link] [9] “Bus Arrival Time Prediction using Bayesian SVR,” 2021. [Link] [10] P. Pov et al., “Bus Arrival Time Prediction Using Machine Learning Ap-proaches,” 2025. [Link] [11] M. Chen, X. Liu, and J. Xia, “Dynamic Bus Arrival Time Prediction UsingKalman Filtering,” 2004. [Link]