In this article, a new method for solving multidimensional population balance (PB) equations for agglomeration is presented. First, discretized 2-dimensional PB (DPB) is derived and by applying probability approach, this method is extended to n-dimensional PB. The discretization of size intervals is in the form of v(j) = 2(j-i) v(i) and the volume based PB is used. Because of conservative nature of volume, all internal coordinates are converted to some independent volumes. A simple and clear procedure for solving multidimensional PB is proposed using probability concept. In addition, various types of birth and death mechanisms involved in PB are analyzed separately. Moreover, results of obtained DPB in 2 and 3-dimensional are compared with time variations of zeroth moment of continuous 1D PB and excellent agreement is found. Finally, the developed method is applied for the simulation of granulation processes and the predicted results are presented and discussed.