The rapid urbanization and growth in the number of automobiles in India have led to increased congestion on roads and a significant impact on human life and the environment. However, the use of intelligent transportation system data in smart-city programs has opened new opportunities for researchers to utilize this data to optimize traffic monitoring, prediction, and forecasting, and to improve policy formulation by governments and societies. This text discusses various approaches for prediction, including statistical and machine-learning techniques, and the mapping of road-level vehicular emissions. A gap is identified in city-specific studies for New Delhi, where a model is proposed to predict the TCI Index for the city using a deep learning time series prediction algorithm. The index is then used to prepare a road-level Carbon monoxide vehicular emission inventory at hourly flux for over eight weeks of the time period. The study offers a unique method for creating high-resolution traffic emission inventories that can be used in many different cities, especially those that face a scarcity of publicly available data. The results of the simulation are presented in graphical form and show that the predicted values accurately identify the trend of actual values and the Carbon Monoxide emission inventory is successfully generated.