Traffic congestion has become a critical issue in modern urban environments due to rapid population growth and the increasing number of vehicles. Traditional traffic management systems rely heavily on real- time monitoring techniques such as surveillance cameras and sensors, which are expensive and reactive in nature. These systems fail to provide predictive insights that can help in proactive traffic management. This paper presents a machine learning-based approach for predicting traffic congestion using historical traffic data. The proposed system, TrafiVista, utilizes supervised learning algorithms such as Random Forest and XGBoost to analyze traffic patterns and classify congestion levels into Low, Medium, and High categories. The system is implemented as a web-based platform using FastAPI, React, and SQLite, allowing users to input parameters such as time and location to obtain predictions efficiently. The approach offers a cost-effective, scalable, and intelligent solution for traffic prediction without relying on expensive infrastructure. By leveraging historical data, the system enables better traffic planning and decision-making, contributing to improved transportation efficiency.

This paper presents a machine learning-based traffic congestion prediction system that utilizes historical traffic data to forecast congestion levels effectively. The proposed system addresses the limitations of traditional traffic management approaches by introducing a predictive model that enables proactive decision-making. By using machine learning algorithms such as Random Forest and XGBoost, the system can accurately classify traffic conditions into different congestion levels based on input parameters like time and location. Overall, the proposed system demonstrates the effectiveness of machine learning in solving real-world traffic problems. It provides a cost-effective and scalable alternative to traditional traffic monitoring systems. By leveraging historical data and predictive analytics, the system contributes to improved traffic planning, reduced congestion, and better transportation management. This work highlights the potential of intelligent systems in transforming modern traffic management solutions.

[1]A. Hazarika, “Edge ML Technique for Smart Traffic Management in Intelligent Transportation Systems,” 2024. [2]A. Saraff, “Indian Traffic Surveillance Video Summarization Using Machine Learning,” 2025. [3]C.-J. Lin and J.-Y. Jhang, “Intelligent Traffic Monitoring System Using Machine Learning Models,” 2022. [4]H. Nashaat, “Machine Learning-Based Traffic Prediction Using Data Reduction Techniques,” 2024. [5]D.-H. Shin, “Prediction of Traffic Congestion Using Machine Learning Approaches,” 2020. [6]Y. Lv, Y. Duan, W. Kang, Z. Li, and F. Wang, “Traffic Flow Prediction With Big Data: A Deep Learning Approach,” 2015. [7]J. Zhang, Y. Zheng, and D. Qi, “Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Prediction,” 2017. [8]X. Ma, Z. Tao, Y. Wang, H. Yu, and Y. Wang, “Long Short-Term Memory Neural Network for Traffic Speed Prediction,” 2015. [9]W. Min and L. Wynter, “Real-Time Road Traffic Prediction With Spatio-Temporal Correlations,” 2011. Z. Chen, X. He, and J. Wang, “Hybrid Machine Learning Model for Traffic Flow Prediction,” 2018.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close