Odor pollution is the biggest source of complaints from citizens concerning environmental issues after noise. Often, the need for corrective actions is evaluated through simulations performed with atmospheric dispersion models. To save resources, air pollution control institutions perform a first-level odor impact assessment, for screening purposes. This is often based on Gaussian Dispersion Models (GDM), which can be executed through user-friendly software that doesn’t need high computational power. However, their outputs tend to be excessively conservative regarding the analyzed situation, rather than representative of the real in-site conditions. Hence, regulations and guidelines adopted at an institutional level for authorization/control purposes, are based on Lagrangian Particle Dispersion Models (LPDM). These grant a more accurate modelling of the pollutants’ dispersion but are very demanding regarding both the needed users’ technical skills and high computing power. The present study aims to increase the accuracy of screening odor impact assessment, by identifying the correlation function of the outputs derived from the two simulation models. The case-study is placed in northern Italy, where a single-point source, with various stack heights, was considered. The identified correlation functions could allow institutions to estimate the results that would have been forecasted with the application of the more complex LPDM, applying, however, the much simpler GDM. This grants an accurate tool which can be used to address citizens’ concerns while saving workforce and technical resources.