The shear capacity of concrete members is a major challenge of structural concrete research through the years. Many theoretical models have been developed and various experiments have been performed, focusing on the accurate prediction of the shear behavior. Many of the available theoretical models assume that a large contribution of the shear capacity is transferred through cracks by a mechanism often recognized as aggregate interlock. At the same time, due to the increase in the complexity of the structures and the development of the concrete technology, the mechanical properties of concrete have improved significantly. This fact leads to the need for modification of the existing models or the development of new ones, accommodated to the improved materials. Since, the aggregate interlock plays a significant role in the development of the shear capacity, the present research proposes a new numerical methodology for the calculation of the aggregate interlock in high strength concrete in which aggregates break, based on the widely recognized model proposed by Walraven and the results of direct surface roughness measurements. The crack surfaces of concrete cylindrical specimens drilled from a 70 years old existing concrete bridge and newly casted cubic specimens generated by splitting tensile tests were measured by a laser scanner. Moreover, the surface of a reinforced deep beam after flexural shear failure was measured as well. The measured crack surfaces were used to implement the plasticity based aggregate interlock model proposed by Walraven with an algorithm which was validated with Walraven’s theoretical model, using a so-called mesostructural model. The output of the analysis gave suggestions on the adjustment of the available aggregate interlock model for high strength concrete. The proposed model is then implemented into a shear test on a 1.2 m concrete beam, which has a concrete strength larger than 70 MPa and the aggregate interlock seems to influence significantly the shear resistance of a cracked section. Based on the observation of the surface roughness of a crack, the thesis further proposed that with a sufficiently large crack face, the localized variation in the crack surface is averaged out. Thus, the surface can be used to develop a master curve for the given concrete type. In the last part of the study, two improvement suggestions are given regarding the Critical Shear Displacement Theory. The one point is relevant to the simplification of the crack profile that can be changed from a straight line into a more inclined and the second point is related to the correction factor considering the fracture of the aggregates, that should be dependent on the crack width.