In urban areas, road traffic is a major source of carcinogenic polycyclic aromatic hydrocarbons (PAH), thus any changes in traffic patterns are expected to affect PAH concentrations in ambient air. Exposure to PAH and other traffic-related air pollutants has often been quantified in a deterministic manner that disregards the various sources of uncertainty in the modelling systems used. In this study, we developed a generic method for handling uncertainty in population exposure models. The method was applied to quantify the uncertainty in population exposure to benzo[a]pyrene (BaP) before and after the implementation of a traffic management intervention. This intervention would affect the movement of vehicles in the studied area and consequently alter traffic emissions, pollutant concentrations and population exposure. Several models, including an emission calculator, a dispersion model and a Geographic Information System were used to quantify the impact of the traffic management intervention. We established four exposure zones defined by distance of residence postcode centroids from major road or intersection. A stochastic method was used to quantify the uncertainty in the population exposure model. The method characterises uncertainty using probability measures and propagates it applying Monte Carlo analysis. The overall model predicted that the traffic management scheme would lead to a minor reduction in mean population exposure to BaP in the studied area. However, the uncertainty associated with the exposure estimates was much larger than this reduction. The proposed method is generic and provides realistic estimates of population exposure to traffic-related pollutants, as well as characterises the uncertainty in these estimates. This method can be used within a decision support tool to evaluate the impact of alternative traffic management policies.