Conflict resolution in hybrid intelligent multiagent systems

The paper discusses an algorithm for reducing the intensity and resolving conflicts that arise in hybrid intelligent multiagent systems. The proposed algorithm is an integral part of the method for managing problem- and process-oriented conflicts in such systems. This method makes it possible to identify decision-making situations, stimulate, if necessary, and subsequently resolve constructive forms of conflict, as well as prevent its destructive forms using the proposed algorithm. Stimulation of the conflict together with the heterogeneity of the system according to the knowledge of specialists simulated by agents and the methods used by them provides an all-sided consideration of the problem posed. The proposed algorithm for reducing the intensity and resolving conflicts allows agents to reconcile automatically their positions on solving the problem and work out a single collective solution in order to save the user from the need for manual analysis and choosing a solution from the array of alternatives proposed by individual agents.