Forecasting behaviour of a stochastic multi-agent system


The method of forecasting macro parameters representing behavior of a stochastic multi-agent system developed to study behavior of operators of complex technical systems is presented. This system is described by the Markov random process with discrete states and time. It introduces the game interaction between a set of agents and a target. The behavior dynamics of such a system is represented in the terms of macro parameters which are considered convenient for interpretation and practical control.