It's provided to use multiagent system's technologies and ontologies for solving problems, directly relating to analyses, diagnostic, evaluation and identification of situations, forecasting of outcomes, information fusion, coordination of unmanned aerial vehicle's group operation. An approach is suggested, which allows analysing the correctness of ontology and its completeness, searching for disturbances in semantic links, agreement of representations about the subject area and the semantics of objects described in terms of different ontologies during their merging. Issues of mapping the elements of the OWL and SWRL-languages to a knowledge base of a decision support system based on the first-order predicate logic are discussed. The key principles of implementing an effective inference algorithm on ontologies for different strategies of search reduction are described. Key words: multiagent system, inference, knowledge base, decision support system, predicate logic, ontology, simulation modeling, knowledge-oriented system, agent approach.
