Since the surface of rice checks is flooded with water for a long time, reducing processes develop in the soil, the redox potential decreases. As a result, a new ecological situation is created in the rice field. At the same time, reserves of oxygen, nitrate nitrogen, iron oxide, sulfates, humus, and total nitrogen are lost. Nitrous iron accumulates in the soil, and sometimes hydrogen sulfide. With poor drainage, hydrogen sulfide is not fully bound by iron oxide compounds and its excess causes the death of rice crops. The functioning of the rice irrigation system and its interaction with the environment cannot be represented in the form of traditional analytical models. The task is complicated by the need to take into account the probabilistic nature of the processes taking place. Field experiments on the rice irrigation system are dangerous and unacceptable. Therefore, it is necessary to develop a model to consider the behavior of the system when external and internal conditions change, taking into account the probabilistic nature of the processes taking place, to perform simulation modeling. Such problems are easier to solve on the basis of the theory of digraphs, which displays feedback in the generated models of systems, which is impossible with the use of mathematical apparatus. The article discusses the construction of a cognitive map of the functioning of the rice irrigation system, an approach has been developed to compile a functional relationship between the parameters characterizing the vertices of the digraph, which will allow quantifying the strength of the influence of one factor on another. An expression is given for the duration of the onset of a certain state of the rice irrigation system. With the help of the compiled model, it is possible to form a quantitative forecast of changes in the indicators of the system, as well as to find various options for influencing the system under study in order to obtain the best option for controlling the functioning of the rice irrigation system.