Fine particulate matter (PM2.5), with a diameter of 2.5 micrometers or smaller, presents a significant health risk due to its ability to penetrate deeply into the lungs and enter the bloodstream. Conventional exposure assessments often overlook critical factors such as individual movement patterns and spatial variability in pollution, leading to less accurate exposure estimates and masking disparities in vulnerable populations. This study introduces an innovative spatial-temporal agent-based modeling (ABM) approach to capture detailed exposure dynamics within urban airsheds, using the Pleasant Run Airshed in Indianapolis, IN, as a case study. By integrating data from 23 PM2.5 sensors, meteorological variables, and land use data, we modeled PM2.5 concentrations over 50 weeks and simulated exposure for 10,000 virtual agents grouped by susceptibility, reflecting varying levels of health vulnerability. Our results reveal marked exposure disparities across sociodemographic groups, with high-susceptibility agents experiencing significantly greater health impacts. The spatial analysis identifies high-exposure zones near industrial areas and transportation corridors, underscoring the urgent need for targeted environmental justice interventions. This study demonstrates ABM’s potential to capture spatial-temporal exposure variability and illuminate inequities in pollutant exposure, offering critical insights for public health policy to reduce environmental health risks. Future research should explore combining ABM with multi-pollutant analysis to comprehensively address complex urban air quality challenges and promote equitable health outcomes.