The study aims to develop an emission inventory (EI) approach and conduct an inventory for vehicular sources in Dhaka City, Bangladesh. A meso-scale modelling approach was adopted for the inventory; the factors that influence the emissions and the magnitude of emission variation were identified and reported on, which was an innovative approach to account emissions unlike the conventional inventory approaches. Two techniques for the emission inventory were applied, viz. (i) a combined top-down and bottom-up approach that considered the total vehicle population and the average diurnal on-road vehicle speed profile in the city and (ii) a bottom-up approach that accounted for road link-specific emissions of the city considering diurnal traffic volume and speed profiles of the respective roads. For the bottom-up approach, road link-specific detailed data were obtained through field survey in 2012, where mid-block traffic count of the day, vehicle speed profile, road network and congestion data were collected principally. The emission variances for the change in transport system characteristics (like change in fuel type, AC usage pattern, increased speed and reduced congestion/stopping) were predicted and analysed in this study; congestion influenced average speed of the vehicles, and fuel types in the vehicles were identified as the major stressors. The study performance was considered reasonable when comparing with the limited number of similar studies conducted earlier. Given the increasing trend of private vehicles each year coupled with increasing traffic congestion, the city is under threat of increased vehicular emissions unless a good management strategy is implemented. Although the inventory is conducted for Dhaka and the result may be important locally, the approach adopted in this research is innovative in nature to be followed for conducting research on other urban transport systems.