WP2: Methodology

Leading Unit: UNIMORE

Given the gap between analysis and design methodologies on the one hand and infrastructures for the implementation on the other hand, this workpackage will aim at the proposal of an innovative methodology that, starting from the current scenarios and needs, allows not only to design agent based applications, but especially provides a concrete support during the whole software life cycle. To this purpose, the methodology will take into account also the existing approaches based on infrastructures that support the implementation and the execution of agent based applications, thus to remove the gap between methodologies and infrastructures. The methodology will be based on the meta-model proposed in WP1. The activities of this workpackage will be organized in four activities as follows.

A2.1: Study of the state of art

The first activity of the workpackage will be dedicated to the study of the state of the art in the field of methodologies for the development of agent based applications. This study will be different from that of A1.1 since it will focus more on the activities and the process of the methodologies, rather than the concepts used in the methodologies. Those methodologies considered "stable" in the field of agent-oriented software engineering will be taken into account, because they have the advantage of having been adopted, tested and sometimes even improved along years. During this study, the more recent methodologies will not be omitted, because they can have innovative aspects, due to the evolution both of the software engineering and of the agent world; moreover, the most recent methodologies can be born from concrete needs discovered in the exploitation of previous methodologies.

Our aim is not to understand which is absolutely the best methodology (supposing that it can exist), but to try to critically emphasise what are the interesting aspects of each methodology, and, most important, what aspects are required to match methodology concepts to infrastructure needs. A criterion to surely take into account will be which life cycle phases are covered by the methodology, while other criteria can emerge from the study of the methodologies themselves.

Expected Results

Scientific papers will be produced, including a critical review of the existing methodologies, still focusing on the aim of our research, that is to bridge the gap between methodologies and infrastructures.

Begin/End

month 1 - month 6

Deliverables

• D2.1 Document describing the state of art