Vol. 325 No. 5 (2014): Информационные технологии

Interactive methods for optimizing the management of a coherent evolution of the organizational and technological system

The relevance of the work is caused by the fact that the information technology effectiveness in the system of controlling evolution of organizational-technological system in the conditions of uncertainty and risk depends first of all on the way the people using the technologies identify and perceive objective evidences of the emerging situations, build assessments, form new knowledge, develop and coordinate behaviors. The main aim of the research is to justify the possibility of using interactive, iterative procedures in coordinated planning tasks for identifying technological reserves and ensuring progressive evolution of organizational and technological systems by maximum complete use of production staff creative potential. Research methods: general methodology and methods of system analysis, simulation and modeling, the methods of theories: fuzzy sets, active systems, decision making, artificial intelligence, fuzzy systems and fuzzy logic, game theory. Results. The authors have investigated the convergence of interactive procedure of the agreed optimization in the task of controlling evolution of organizational-technological system, consisting of a centre and m agents at different types of center awareness on agent abilities. In the case of complete centre awareness it was shown that the solution to the problem of optimal coordinated planning will be implemented by the agents, if stimulus and the plan are found under which the specific value of the purposeful state situation will exceed the value of the same indicator for all other variants. For the case of incomplete centre awareness it was shown that the convergence to the optimal solution will be provided by additional agent stimulating for a more complete disclosure of their capabilities. The paper introduces the iterative procedure using the idea of Polterovich block programming algorithm. The author formulated the condition for its convergence. For the case of center incomplete awareness on agent preferences it was shown that the solution can be obtained applying an iterative algorithm which uses a relaxation scheme of conditional gradient. The paper introduces the conditions under which one can build the procedure of the interactive information exchange between the center and the agents that guarantee the convergence of the solution of coordinated optimization task. The authors carried out the experimental validation of the approach proposed in the development of the production program for producing refrigerants and fluoropolymers.

Keywords:

active system, intelligent agent, consistent optimization, decision making, interactive cooperation

Authors:

Gennady Vinogradov