Beyond the estimation of daily exposure profiles of agents, agent-based models can be used to investigate the role of different urban areas in noise exposure. This study explores this potential by proposing a new approach to spatio-temporal mapping of the accessibility to quiet areas in agglomerations, integrating mobility into the analysis of place effects. Using an agent and activity-based framework, composed of EQASim, MATSim and NoiseModelling, the Nantes Metropolitan Area (France) is modeled. The definition of the quiet areas conditions and the accessibility metric are analyzed in order to understand the influence of these parameters on the identification of critical urban areas where agents don't have their access to quiet areas assured. The results show a great variability of accessibility according to its definition and emphasize the need to include new variables in the assessment.