Portal de Periódicos da CAPES

Towards a More Expressive and Refinable Multiagent System Engineering Methodology

2004; Springer Science+Business Media; Linguagem: Inglês

10.1007/978-3-540-25943-5_9

1611-3349

Shiva Vafadar, Ahmad Abdollahzadeh Barfouroush, Mohammad Reza Ayatollahzadeh Shirazi,

AI-based Problem Solving and Planning

In this paper, we improve and extend the MaSE methodology to bridge the gaps in this methodology. First, we propose a methodology improvement process and, based on this process, we report the discovered gaps and weaknesses in the methodology. For removing the reported weaknesses, we introduce the "Role Schema" to document roles properties and the "Knowledge Modeling" step in order to model knowledge of each single agent in the analysis phase of the methodology. We also propose the "Agent–Object model" to decrease design and implementation complexity and improve efficiency of the developed agent-based system. In the improvement process, for evaluating the proposed refinements and extensions we have analyzed and designed the CASBA multiagent system with the improved MaSE. We will show that these improvements will increase expressiveness and refinability of the methodology and maintainability of the developed agent-based system.