Cake filtration is a widely used solid-liquid separation process. However, the high flow resistance of the nanoporous filter cake lowers the efficiency of the process significantly. The structure and thus the permeability of the filter cakes depend on the compressive load acting on the particles, the particles size, and the agglomeration of the particles. The latter is determined by the particle charge and the ionic strength of the suspension, as described by the Derjaguin-Landau-Verwey-Overbeek (DLVO) theory. In this paper, we propose a combined stochastic rotation dynamics (SRD) and molecular dynamics (MD) methodology to simulate the cake formation. The simulations give further insight into the dependency of the filter cakes' structure on the agglomeration of the particles, which cannot be accessed experimentally. The permeability, as investigated with lattice Boltzmann (LB) simulations of flow through the discretized cake, depends on the particle size and porosity, and thus on the agglomeration of the particles. Our results agree qualitatively with experimental data obtained from colloidal boehmite suspensions.