Рurpose of research. Increasing the efficiency of cargo port business processes through the use of risk management technology based on the application of a cognitive modeling.Methods. A cognitive modeling of cargo port risk management is proposed, based on a comprehensive step-by-step application of the concept of multi-level goal setting, which involves a thorough elaboration of the cargo port's goals, as well as indicators for assessing the achievability of goals by developing a balanced scorecard (BSS) and constructing a logicalprobabilistic (LP) model; a logical-ontological model developed on the basis of the connections established by the LP model; a simulation model used to check recommendations for adjusting the elements of the system under consideration, developed on the basis of queries to the ontological model, in order to select the most acceptable options for recommendations or combinations thereof and formulate management decisions based on them. Results. Based on the formulated purpose of the study and the assigned tasks, a concept of cognitive modeling was developed, which involves the use of knowledge about the connections between risks, goals, indicators for assessing port activities, as well as clarifying coefficients and the nature of their influence on each other in order to develop recommendations for managing the risks of a cargo port at based on queries to the ontological model. The simulation model within the framework of the proposed conceptual approach allows us to develop management decisions on adjusting the operational components of the system in order to prevent risk situations in the long term (at the tactical and strategic levels) taking into account the influence of external factors. Cognitive modeling is based in this work on the integration of logical-probabilistic, logical-ontological and simulation modeling. Conclusion. As a result of the implementation of the set goals and objectives, a cognitive model of cargo port risk management was proposed. This model combines various types of modeling and takes into account different levels of management. As a result of experiments with a simulation model, the most effective recommendations generated based on a query to the ontological model are selected as management decisions.
