Article 5

Authors

Ram Kishor Singh, Research Scholar, Department of Civil and Environmental Engineering, Birla Institute of Technology, Ranchi, Jharkhand

Bindhu Lal, Professor, Department of Civil and Environmental Engineering Birla Institute of Technology, Ranchi, Jharkhand

Navin Prasad, Research Scholar, Department of Civil and Environmental Engineering, Birla Institute of Technology, Ranchi, Jharkhand

Vehicular air pollution is a significant contributor to deteriorating urban air quality, posing severe health and environmental challenges. In this study, Classification and Regression Tree (CART) modelling is used to analyse and predict vehicular emissions for particulate matter (PM10), nitrogen oxides (NOx), and sulphur oxides (SOX). The model’s results demonstrate high accuracy in predicting pollution levels, highlighting specific conditions and locations prone to elevated emissions. The best predictions for SOX and NOX, are those with R² values of 0.61 and 0.66, respectively, while the best results for PM10, are those achieving an R² value of 0.51. The IA is highest for NOX (0.81) and lowest for PM10 (0.576). RMSE (1.53) is the least for SOX. This research underscores the utility of Classification and Regression Trees (CART) modelling in supporting data-driven urban planning and environmental management. By accurately predicting and analysing vehicular pollution, the study provides a robust framework for policymakers to design targeted interventions to improve air quality and minimise public health risks.