Supercritical water hydrothermal synthesis of bismuth oxide nanoparticles was investigated, using Response Surface Methodology and numerical simulation by population balance equation, aiming to optimize variables and minimize the tests. Pure crystalline tetragonal Bi2O3 nanoparticles were prepared only at three operating conditions. Particle size and distribution (PSD) predicted, compared well with experimental results; thereby confirming the model accuracy. Theoretical size and PSD, determined for twelve other proposed operating conditions, showed respectable fit for the PBE modeling and second-order polynomial models (R-2 = 0.9948 and 0.9960, respectively). Optimal conditions (T = 409.2 degrees C, C = 0.3 M and pH = 10.0) were verified empirically, justifying further validity/adequacy of the approach adopted here.