Розподілена система підтримки прийняття групових рішень

Abstract

The decision-making process is an integral part of modern society. Research shows that over the last decades, the use of groups within organizations has increased dramatically, and each organization operates based on decisions made by groups of individuals within the organization. Almost every aspect of social, political, legislative and economic life functions through the decision making of groups of individuals. Systems theory views group decision making as interdependent forces capable of being analyzed from the perspective of other forces. The behavior of individuals in groups is due to many reasons, and the systematic method emphasizes multiple cause and effect relationships and the complex interplay of forces. Effective decision-making is seen as a natural consequence of the ability of decision-makers to analyze and understand the group decision-making process, but no one can acquire this ability without actively participating in the group decisionmaking process. A group decision, in general, is the choice of group members from among the alternatives available to them. Very rarely does a group of individuals make decision in isolation from the rest of the group. There are currently no simple tools and systems available to make the best decisions, which raises the need for research in this field. Group decision support systems are interactive computerized systems that help a group of decision-makers use data and models to identify and solve tasks. In this article, the use of Bayesian network analytics is proposed as an approach to implement a distributed group decision support system and resolving conflict situations in the decision-making process. For this purpose, the information technology of the distributed group decision support system was developed, the information model and dynamic data structures were proposed, and the software implementation of the group decision support system was developed. Experimental studies of the developed system have shown that by using Bayesian networks one can avoid conflict situations caused by the use of majority principle and its various modifications.