Relative permeability and capillary pressure saturation functions are essential to Reservoir Engineering because they determine the efficiency of water-flooding operations on the microscopic and macroscopic scale. These functions are required for predicting the reservoir performance through the whole reservoir life time. Generally, the relative fluid-phase permeability in the formation rock can be measured by performing displacement experiments in core sample by either steady state or unsteady state flooding experiments. Conventional analytical interpretation of laboratory SCAL experiments as performed by many service laboratories may add uncertainty to relative permeability and capillary pressure data and consequently to reservoir simulation. Relative permeability and capillary pressure function can more reliably be obtained by numerical history matching displacement experiments. The main problems of the analytical approach are the crude approximations behind the interpretation models such as the JBN approach [1]. This is the restrictive assumption that neglects action of capillary forces, which is especially a problem close to residual oil saturation. Thus numerical modelling of SS (steady-state), USS (unsteady state) and C (centrifuge) experiments and more specifically history matching of related production, pressure and saturation data are the way to obtain more accurate results because full physics is taken into account. Finally, the procedure will be verified by comparing the results obtained in this study to literature data obtained by using different simulation tools and approaches. The proposed master thesis aims on numerical interpretation of SCAL (Special Core Analysis) data in order to improve the quality of relative permeability and capillary pressure saturation functions. By simulating and history matching SCAL experiments we will overcome typical experimental issues and the deficiencies of analytical interpretation methods. In the frame of the thesis, numerical models for different SCAL techniques will be setup and experimental data will be described. To achieve the best interpretation and for better handling, the history matching procedure will be automated.