The following paper presents the design and implementation of a congestion control to a data network. First make a contextualization of the problem of congestion in data networks and how this phenomenon affects the performance of the network. Then it shows which have been the main ways to control the phenomenon of management, most of the solutions are based on the modification of existing protocols, such as TCP, but these modifications to the protocols involve a series of adaptations of the different elements network. Using artificial intelligence tools, in this case a fuzzy controller, is designed the control model, which works by an allocation of the bandwidth of the ports of a switch (emph switch). To check the operation of the model is implemented a controlled environment that emulates the characteristics of the corporate LAN National University. Shows the results, where we see the model performance and versatility of the controlled environment.