The city of Milan, located in the Po basin, is often affected by episodes of poor air quality with negative consequences for the health of the local population. A recent study shows that road traffic plays an important role, being responsible for 72% of NO2 and 28% of the average annual PM2.5 concentrations in the city center (Pepe, et al., 2019). The ability to simulate pollutant concentrations on a local scale, with high accuracy, is a fundamental premise for the evaluation of mobility policies aimed at reducing air pollution. So far, a hybrid modeling system (HMS), based on the Eulerian CAMx model and the Lagrangian particle model AUSTAL2000, has been developed and used to simulate pollutant concentrations in the center of Milan (Pepe, et al., 2016). Although this approach is very promising, it leads to problems of inconsistency due to offline linking between models and double counting of emissions. To address this problem, the Linear Plume in Grid scheme was developed, within the CAMx code, as an extension of the functionality of the PiG scheme (Ramboll Environ, 2016). What is known as the CAMx-LPiG model allows roads to be modeled as linear sources and could be a promising modeling alternative. LPiG is designed to be fully consistent with the Eulerian model and exploits the chemical scheme available in CAMx; therefore, it is able to simulate even reactive chemical compounds in a detailed and coherent manner. This work shows the concentration fields of NOx, NO2, PEC and PM2.5 calculated with the CAMx-LPiG scheme and compares them with the observed data. The preliminary case study modeled a subset of streets in the center of Milan, with 50 m horizontal resolution. CAMx-LPiG provides reasonable results, successfully capturing the spatial variability of the concentration fields due to the road network. Furthermore, compared to the HMS, this new scheme allows greater flexibility of implementation, depending on the mobility policy we need to evaluate.