Analysis and Design of Agent-Oriented Information Systems

Analysis and design of information systems (IS) is the process of eliciting the system’s requirements and transforming them into a model that could be used to develop IS. Analysis and design of agent-oriented information systems (AOIS) relates to the very same process using the multi-agent paradigm. A comprehensive and rigorous methodology for developing multi-agent systems is lacking [Elammari and Lalonde 1999, Odell et al. 2000]. Most existing multi-agent systems were developed in an ad-hoc manner, and systems developers paid little attention to requirements specification and the analysis process [Treur 1999a]. In this paper, we describe different methodologies that are suitable for analyzing and designing AOIS. We view the analysis and design process as a modelling problem, and look at the representation models when studying the frameworks. In addition to agent-oriented Analysis and Design (A&D) methodologies, we also review the representational component of agent-oriented architectures, languages, and development frameworks. These neighboring fields provide valuable insight of what should be captured by models of an agent-oriented A&D methodology. To better understand the various approaches, we analyze their strengths and weaknesses, classify them according to two dimensions. This then enables us to compile a list of requirements for an ideal unified agent-oriented modelling methodology. Categories and Subject Descriptions: D.2.1 [Software Engineering]. Requirements/Specifications – Languages and Methodologies; D.2.10 [Software Engineering]. Design – Methodologies and Representation; H.4.m [Information Systems Applications]: Miscellaneous; I.2.1 [Artificial Intelligence]: Applications and Expert Systems; I.2.11 [Artificial Intelligence]: Distributed Artificial Intelligence – Languages and structures; K.6.1 [Management of Computing and Information Systems]: Project and People Management – Systems analysis and design