Modeling and estimation of cooperative index for multi-agent systems using execution graph

Abstract

A multi-agent system (MAS) is a system composed of multiple interacting intelligent agents. MAS can be used to solve problems that are difficult or impossible for an individual agent to solve. The different characteristics of MAS help in solving highly complex distributed problems. One of the important characteristics of MAS is its cooperative nature. This character helps different agents to interact with each other by exchanging messages. One of the major challenges in MAS is quantifying the cooperation between agents. In this paper, we propose a new methodology to compute the cooperative index of MAS in the early stages of its development. For this calculation, execution graph is used to model the software specifications. The proposed methodology is illustrated with a case study and the results are compared with the cooperative index obtained from the Unified Modeling Language (UML) sequence diagram. © 2013 Springer.