The rapid increase in vehicles because of urbanization has created chaotic transportation networks causing an excessive delay in travel, consumption of fuel, and air pollution, becoming one of the major issues across the globe in urban places. Therefore, affecting mobility, safety, and quality of life. Addressing this pressing issue, this paper develops a dual ensemble regressor model and for predicting traffic management based on machine learning in order to enhance traffic predicting accuracy. Furthermore, our paper compares the results of ensemble model with the other widely used regression models including Random Forest, CatBoost, Gradient Boosting, K-Nearest Neighbours, Ada Boost, Support Vector Regression, Linear Regression. The models have been employed to a Kaggle dataset containing 33,707 features and 15 instances. This paper also underscores the past research literatures that have approached this serious issue. The results show the superiority of ensemble modelling approach greatly improving the predictability of traffic volume, yielding high accuracy metrics including low Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and high R-squared values. The proposed model gives the RMSE and MAE as low as 342.50 and 205.47, respectively, and an impressive R-squared value of 97%, greater than all the other models compared. This work aims to add to the existing knowledge body on traffic prediction and traffic management authorities, to overcome congestion and promote efficiency within transportation systems, while helping to mitigate negative environmental impact.