The diploma thesis deals with the comparison of possible concepts of solving carousel fraud on value added tax. The goal is to identify the most appropriate model which leads to the prevention of carousel fraud within the European Union. The considered solutions are divided into two groups, the technological and structural concept. According to the research, it was found that the most suitable of all technological solutions seems to be the VATCoin model and also the solution using RTvat. In case of the structural solutions, the reverse charge model and Vanistendael’s foreign branch apportionment are identified as the most suitable models. Subsequently, in the work, the most suitable solutions are compared, regardless of their classification...