The method of construction of traffic flow control models using the controlled networks theory is considered. A detailed example of construction of the model is presented. The obtained model is a system of nonlinear finite-difference equations. The model can include any number of intersections and considers different features of the road network. To solve the problem of optimal control of traffic flow a genetic algorithm is applied. Then using methods of parametric identification we find the parameters of flow distribution and the capacity of manoeuvres at intersections. © 2012 IFAC.