Network Traffic Analysis and Bandwidth Forecasting for using Meta’s Prophet: A Case Study

Abstract

This study created a forward-looking bandwidth prediction system for students’ halls of residence at Landmark University. The system uses Meta’s Prophet, a method for analyzing patterns in data over time, and was trained on past internet traffic data from October to December 2024. The system was able to predict future bandwidth usage with over 90% accuracy. To assess how well the system worked, several common metrics were used, including mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE). The MAE was calculated as 10,099,863.10 bits per second (bps), and the RMSE was 13,570,959.58 bps. While the mean squared error (MSE) appears large numerically, this is anticipated due to the size of the bandwidth data involved in its calculation. Importantly, the prediction errors are considered reasonable when considered in relation to the actual peak bandwidth usage, which fluctuated between 47 and 50 megabits per second (Mbps). These findings suggest that machine learning can be a valuable tool for refining network infrastructure and improving the user experience quality of service (QoS) in environments with many users, such as university residences.