MAS Ontology: uma ontologia de métodos orientados a agentes

Abstract

The agent-oriented modeling emerges as a paradigm in software development, considering the amount of initiatives and studies that refer to the use of software agents as a solution to address more complex problems. Despite the popularity of using agents, experts bump in the lack of universality of a methodology for the construction of Multiagent Systems (MAS), as they end up sinning by excess or lack of solutions to model the problem. This thesis proposes the use of an Ontology based in Methodologies for Multi-Agent Systems, following the principles of Situational Method Engineering, which proposes to use fragments of methods for constructing methodologies based on the specificity of the development project. The aim of this work is to consolidate the knowledge in the area of Multiagent Methodologies by helping the software engineering to choose the best methodology or the best fragment method able to model a specific Multiagent System.