In the present paper, a system dynamic model is presented to predict the final particle size of milled powder during ball milling process. The presented model is used to obtain the optimum ball size, milling speed and milling time that achieve the best particle size reduction of metal-matrix nanocomposites. Parametric study is performed using the presented analytical model to study the influence of ball size and milling speed on the milling efficiency. The predictions of the presented model are validated with experimental results done during this work for Cu-5% ZrO2 nanocomposite and others available in the literature. The results show that the milling time required to achieve the steady state condition for Cu-5% ZrO2 nanocomposite is 15 h. At 15 h of milling, ZrO2 particles are highly uniform distributed in Cu matrix and the microhardness is increased from 75.4 HV for Cu to 197.6 HV for Cu-5% ZrO2 nanocomposite. After 15 h, the particle size reduction rate is too low and the hardness improvement rate is too low as well (204.1 HV after 20 h milling) which make the milling process after 15 h is not appreciable. (C) 2018 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights reserved.