Machine Learning-Based Traffic Congestion Prediction and Policy Simulation for Sustainable Urban Transportation in Bengaluru

• Author(s): Vedanth Parida; Prof. Rakshitha B. S
• Paper ID: 1717886
• Page: 2511-2522
• Published Date: 19-05-2026
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 9 Issue 11 May-2026

Abstract

Traffic congestion has become a major urban challenge affecting travel time, fuel consumption, road safety, commuter productivity, and environmental sustainability. Rapid growth in private vehicle usage in Bengaluru has increased pressure on urban road networks, especially in highly active commercial, residential, and employment corridors. This study proposes a machine learning-based framework for traffic con-gestion prediction and simulation-based evaluation of urban transport policies using Bengaluru traffic datasets. The work includes data cleaning, exploratory data analysis, outlier inspection, feature engineering, model training, policy simulation, residual validation, and explainable artificial intelligence analysis. Three machine learning models, namely Linear Regression, Random Forest Regressor, and XGBoost Regressor, were im-plemented and evaluated for congestion prediction. Engineered features such as Bus Lane Impact, Transit Efficiency, Carpool Efficiency, and Fuel Wastage Index were developed to support policy-aware analysis. Experimental results showed that XGBoost achieved the best performance with an R2 score of approximately 0.989, outperforming Linear Regression and Random Forest. SHAP-based explainability analysis showed that Traffic Volume was the most influential factor in congestion prediction, while bus lane impact, transit efficiency, and carpooling-related features contributed to policy interpretation. The proposed framework demonstrates how machine learning can support intelligent traffic planning and simulation-based policy evaluation for sustainable urban mobility.

Keywords

Traffic Congestion Prediction, Machine Learning, XGBoost, Random Forest, Policy Simulation, Bus Lane Impact, Carpooling, Explainable AI, Bengaluru Traffic

Citations

IRE Journals:
Vedanth Parida, Prof. Rakshitha B. S "Machine Learning-Based Traffic Congestion Prediction and Policy Simulation for Sustainable Urban Transportation in Bengaluru" Iconic Research And Engineering Journals Volume 9 Issue 11 2026 Page 2511-2522 https://doi.org/10.64388/IREV9I11-1717886

IEEE:
Vedanth Parida, Prof. Rakshitha B. S "Machine Learning-Based Traffic Congestion Prediction and Policy Simulation for Sustainable Urban Transportation in Bengaluru" Iconic Research And Engineering Journals, 9(11) https://doi.org/10.64388/IREV9I11-1717886