Air pollution, as one of the serious environmental challenges, has significant negative impacts on public health and quality of life. This study investigates and models air pollution using the traffic density index and wind speed. The main objective of this research is to analyze the impact of various factors on air pollution and provide reliable predictions for pollutants including carbon monoxide, particulate matter, and sulfur dioxide. Modeling was performed using the traffic density index and wind speed. In the data preparation stage, to investigate short-term and cumulative effects, data were organized and analyzed in the form of hourly intervals, as well as two-hour and three-hour averages and sums. The results of this section indicated that using hourly data provides better performance in predicting air pollution compared to cumulative states. By analyzing density data collected on an hourly basis and utilizing neural network and machine learning models, accurate predictions for pollutants were provided. The results showed that air pollution can be effectively modeled using the traffic density index, and the XGBoost model demonstrated the best performance in this regard. This research highlights the importance of using traffic data and meteorological variations in air pollution analysis and the high potential of machine learning models in predicting pollutant concentrations.