In wastewater treatment plants (WWTPs) the production of nitrite as an intermediate in the biological nutrient removal (BNR) process has been widely observed, but not been taken into account by most of the conventional activated sludge models yet. This work aims to develop a mechanistic mathematical model to evaluate the BNR process after resolving such a problem. A mathematical model is developed based on the Activated Sludge Model No.3 (ASM3) and the EAWAG Bio-P model with an incorporation of the two-step nitrification-denitrification, the anoxic P uptake, and the associated two-step denitrification by phosphorus accumulating organisms. The database used for simulations originates from a full-scale BNR municipal wastewater treatment plant. The influent wastewater composition is characterized using batch tests. Model predictions are compared with the measured concentrations of chemical oxygen demand (COD), NH4+-N, NO2--N, NO3--N, PO43--P, and mixed liquid volatile suspended solids. Simulation results indicate that the calibrated model is capable of predicting the microbial growth, COD removal, nitrification and denitrification, as well as aerobic and anoxic P removal. Thus, this model can be used to evaluate and simulate full-scale BNR activated sludge WWTPs. (C) 2009 American Institute of Chemical Engineers AIChE J, 56: 1626-1638, 2010