Search

The article is devoted to the application of discrete-time controlled dynamical economic and mathematical models of production systemsfor solving the problem of constructing and minimax estimation of its predictive set in the fixed time period (the reachability set of the system in the fixed time period). On a given integer time interval, an economic and mathematical model of a controlled production systemthe output of products in the presence of a uncertain demand (controlled dynamical system) is considered, which is described by a system of linear discrete-time recurrent equations. In each time period, the values of the control action are bounded by the corresponding nonempty finite collection of vectors in corresponding finite-dimensional vector space, and the values of the phase vectors and vectors of demand of the dynamical system are bounded by the nonempty convex polyhedrons in corresponding finite-dimensional vector spaces. In the article describes the formalization of the problem of constructing and minimax estimation of the predictive set in the fixed time period of the dynamical system under consideration, as well as the proposed method for its solution. The results obtained in the article can be used to constructing intelligent computer systems for information support in solving actual technical, economic and medical problems. [ABSTRACT FROM AUTHOR]

This article discusses the processes of predicting the state and optimizing the management of the production potential of the region in the presence of the possibility of using a hierarchical agent-oriented control system. It is proposed to use a deterministic economic and mathematical model for the processes under consideration, in which the dynamics of the basic factors (phase vectors) characterizing the state (potential) of industrial objects of the region as a whole, the municipalities that form it and specific industrial enterprises with a three-level spatial structure, is described by the corresponding linear discrete-time vector-matrix recurrent equations in the presence of control actions (controls). In the proposed agent-oriented control system, three levels of management decision-making are distinguished – the dominant (regional or first level of management), which is at the disposal of a generalized regional agent, the first subordinate (municipal or second level of management), which is at the disposal of municipal agents, and the second subordinate (the level of manufacturing enterprises or the third level of management), which is at the disposal of production agents. The control system is hierarchical, in which the choice of control of a generalized regional agent determines the resource capabilities of management of municipal and production agents, and the choice of control of a specific municipal agent determines the resource capabilities of control of the corresponding production agents. All agents of the proposed control system are a priori united by certain information and management links. The industrial potential of the region as a whole, its constituent municipalities and industrial enterprises is estimated by the corresponding convex objective function (objective functional) and all agents of the control system are interested in their minimum values. The paper presents formalizations and general scheme for solving the problems of predicting the state and optimizing the management of the region's production potential. The results obtained in the article can be used in the development and creation of intelligent computer systems for information support and management decision support for regional administrations. [ABSTRACT FROM AUTHOR]