Membrane fouling represents one of the major issues for a membrane bioreactor (MBR). Membrane fouling and high aeration requirements (for inducing shear stress to limit fouling) make MBR operation economically demanding due to high energy costs. Although several studies on MBR fouling have been performed, comprehensive knowledge on how to reduce membrane fouling and consequently save energy is still lacking. An integrated mathematical model for MBR is applied to a University of Cape Town membrane bioreactor with the final aim to reduce the energy costs. In particular, the influence of the aeration intensity, the duration of filtration/backwashing cycles, and the number of membrane cleanings are investigated. Five scenarios are analyzed and compared, each implementing different operating conditions. The features of the analyzed scenarios are quantified by employing Monte Carlo simulations and performance indices partially drawn from literature. The results provide insights about the role played by the main physical/chemical/biological processes in view of a system optimization. As expected, MBR operation at low air flow rate (qa) leads to a substantial reduction of the operational costs (specifically, 20% with respect to the suggested manufacturers ones in terms of qa). Despite such a reduction of qa, a good effluent quality is also obtained as an effect of a high biological cake thickness. Results also show that the values of filtration time (TO higher than those suggested by manufacturers (e.g., Tf=9 min) can be used to increase effluent quality. This study demonstrates that both energy savings and effluent quality can be improved by varying the operational variables with respect to those of the suggested manufacture. One of the main insights gained from this study is that the values of the operating variables (i.e., qa, Tb and TO suggested by the manufactures can be changed to obtain a system that still respects high effluent quality and is characterized by lower economical cost. The proposed modeling approach can be an useful tool for the optimization of the operating conditions in order to reduce the operational costs for MBR systems. (C) 2013 Elsevier B.V. All rights reserved.