Despite the high number of advantages of membrane bioreactors (MBR), operational costs still remain as one of the main obstacles of this technology. In that sense, a full-scale MBR was studied in order to explore and identify viable optimization strategies for improving effluent quality and reducing operational costs through a model-based approach. Opportunities for optimization strategies, identified through the initial experimental phase, were focused on improving the nitrogen removal efficiencies and reducing the aeration energy costs through the aerobic DO set point and recirculation modifications. A mechanistic model was developed to reproduce the operation of the full-scale MBR and predict the effects of the optimization actions. Moreover, a qualitative risk model was also applied to ensure that the DO set point reduction did not negatively affect the microbiology of the activated sludge. The best viable control scenario was identified and then tested in the full-scale MBR. The results achieved maximum improvements on the nitrogen removal efficiencies of 27% and reduction on the aeration energy (7%) without affecting the sludge properties or the filtration performance. The model approached proved in this study and the optimization strategies identified can be generalized for municipal MBRs. (C) 2015 Elsevier B.V. All rights reserved.