The main goal of this project is to verify the numerical model SWASH when dealing with overtopping prediction by comparison of full scale measurement results to results of numerical model testing. For that purpose, the harbour in Zeebrugge (Belgium) has been considered; there, several storm events have been measured in a rubble mound breakwater section. In order to get a proper reproduction of these, a stepwise process has been carried out: First, wave propagation has been tested in SWASH along a flat flume. Results have shown good performance of the software. The next step was to introduce a smooth impermeable dike on a shallow scenario. Also in this case, good performance has been observed; propagating waves reaching the dike are affected from reflection and they are able to run-up the slope, getting overtopping records in some cases. Finally, to get a proper reproduction for the real case in Zeebrugge, porosity was added to the dike in the prior step. It has been concluded that SWASH is not able to model breakwaters only by means of a porous structure; therefore, an impermeable core is required to allow simulation of real phenomena. After conducting some tests over an impermeable core with an outer porous layer, it has been proved the software is not able to reach reliable levels for waves running up the slope. Strong dissipation is taking place within the porous layer, so that SWASH is concluded to underestimate overtopping prediction for rubble mound breakwaters.