Study on Public Bus Ridership Prediction Using Machine Learning: A Case Study of Public Bus in Phnom Penh City

抄録

The public bus system plays a crucial role in enhancing urban mobility and optimizing transportation networks. However, accurately predicting ridership remains a challenge, hindering efforts to improve operations and service quality. This study aims to apply machine learning (ML) techniques to predict public bus ridership in Phnom Penh, leveraging historical ridership data, weather conditions, and temporal features (e.g., day of the week) to improve the accuracy of the predictive model. This study assessed four predictive models: linear regression (LR), random forest (RF), eXtreme gradient boosting (XGBoost), and artificial neural networks (ANNs). The results demonstrate that the XGBoost model emerges as the most effective in capturing complex ridership patterns, based on performance metrics such as mean absolute error (MAE) and root mean square error (RMSE). This research highlights the importance of ML in supporting data-driven decision-making for public transportation systems in Cambodia, contributing to more efficient and sustainable urban mobility solutions.