The process of driving a train was represented in the form of fuzzy situations, given in a table. The conformity between all possible situations and a set of driving decisions was established. The table size is determined by the number of situations which, in turn, depends on the degree of concretization of values. An algorithm of actions of a locomotive driver when driving a train is presented in the form of fuzzy probabilistic graph. Fuzzy numbers, the values of which are recorded in the matrix graph, represent the weights of transitions between vertices. The choice of decision by a locomotive decision support system (DSS) is carried out using the utility criterion. The training system is implemented with the use of the fuzzy classifier that represents fuzzy knowledge base, the input of which receives signals about current state of the traction rolling stock and of the environment. The model of dynamic knowledge base was obtained. As a result of analysis of existing types of intelligent systems, hierarchies, and algorithms of their work, taking into account the working conditions of locomotive crews and railway transport as a whole, the parameters for locomotive DSS were developed. We defined the minimal time it takes for a locomotive driver to make a decision about driving a train and to identify emergency situations. The functions of person that directly affect the efficiency and safety of the locomotive and require support using the intelligent systems were determined. The results of the work allow implementing intelligent DSS in modern locomotives. This will enhance the level of safety and efficiency of driving a train.