Modeling the Evolution of Companies using Intelligent Software Agents Architecture

Abstract

The paper presents the concept of multi agent system that models the evolution of a company. The opportunity of such an approach and the limits of mathematical modeling are presented. The main players on the market are modeled as cognitive, adaptive, heterogeneous agents and evolve in a dynamic environment. The purpose is to use the model of operational agent that has these characteristics. This model is based on using an ATN (Augmented Transition Network) to adapt the behavior of its agent to the changes it detects in its environment. Each agent has an inference mechanism for the first order reasoning. The agents communicate between them through messages and will be implemented in a non synchronized object environment.