The article proposes a method of hybrid construction and support of data transmission routes in programmable means of mobile radio communication based on neural networks. Taking into account the peculiarities of the operation of mobile radio communication networks and means of communication, the article proposes to intellectualize the processes of managing data flows by using knowledge processing technologies in the construction of a control system. The construction of the control system is based on the use of neural networks, the methodology proposes three main steps of its construction, namely the method of hybrid construction of data transmission routes, the method of maintaining data transmission routes and modeling the process of construction and maintenance of data transmission routes in the network using neural networks. Unlike similar methods, which: are used in stationary communication networks with a small number of end nodes, without the possibility of monitoring the functions of data transmission routes; which are characterized by the absence of a clear decision-making hierarchy and the presence of unpredictability when applied in data transmission systems of mobile radio communication networks; do not take into account the peculiarities of the functioning of mobile radio communications. The proposed technique is able to build data transmission routes, increases the time of network operation and reduces the loading of network channels based on the application of a hybrid structure of neural network construction, by minimizing service traffic in the information direction; selection of the type, number and method of control of data transmission routes; introduction of the hierarchy of the decision-making process for the construction of the data transmission route and the selection of data parameters, taking into account the characteristic features of the use of programmable mobile communication tools to ensure the given quality of network service under different conditions of operation of mobile radio communication tools, using neural networks.