The standard Lattice Boltzmann (LB) method is a rather new approach to analyze and simulate fluid flow. This method employs a uniform computational grid for the whole physical domain of interest. In the case of simulating complex flows, and to achieve more accuracy, the system’s gird resolution has to be much finer. This leads to an exponential increase in computational cost, which is directly related to the ratio of the lattice space in any grid system. Therefore, advanced computational methods to reduce such grid-based costs are being interested. MultiGrid (MG) methods are a prime source of important advances in algorithm efficiency, being widely used by many researchers. Unlike other known methods, MG offers the possibility of solving problems with N unknowns and O(N) number of instructions and storage demand for large justify; MARGIN: 0in 0in 10pt" histogram of the particle’s situation. Considering the fact that LB equations are non-linear, non-linear MG equations are also required needed. Therefore the Governing equations of the MG and LB methods are derived. In non-linear solutions an appropriate initial value has to be chosen. The nested algorithm for this requirement is therefore applied. In the section on the results, the simulation of fluid flow in lid-driven cavity is presented as an original LB modeling. Then the results are validated with those in the literature. The derivation of the reformed non-linear MG method equations on the basis of LB equations is also presented. In these equations the residual is restricted in addition to the value of the function. Then the parallel processing algorithm is applied. The effects of several parameters on the precision and the speed of the numerical simulation are investigated. The quality of the results are compared and discussed. As a result of combining LB, MG and parallel processing schemes, a speed-up of more than 150 is achieved compared to an original single grid, single core LB implementation. An example on parallel programming using MPI is presented in the appendix. The innovation of this thesis is an explicit combination of the LB and MG method by employing parallel processing architectures to increase the computational performance. Key Words : Lattice Boltzmann, MultiGrid, Computational performance, Parallel processing