Poster session only! This paper is a collaborative effort with Mitel. It has self-modifying systems as its central theme. It attempts to show, by argument and example, that use case maps are uniquely suited to understanding and describing the kind of self-modifying agent systems with which Mitel is currently experimenting. However, the explanations were too terse and the link to agent systems not strong enough for the reviewers. At the time, it was submitted, it stood as our best effort in this direction.

Abstract

Agents are appropriate for defining, creating, maintaining, and operating the software of distributed systems in a flexible manner, independent of service location and technology. However, we humans have difficulty understanding or defining how a system of agents works as a whole to accomplish some application purpose when the only models we have to work with are provided by a programming language or a CASE tool that supports a set of relatively low level software design diagrams, such as object interaction diagrams, and class inheritance hierarchies. We need a macroscopic, system-oriented model to provide a means of both visualizing the behaviour of systems of agents and defining how the behaviour will be achieved, at a level above such details. Use case maps provide such a model. This paper outlines principles of applying use case maps to help with the process of discovering agents and their relationships, defining abstract system models of sets of collaborating agents, constructing implementable system definitions from the models, arranging for the agents to instantiate the systems from the definitions, and evolving the system models and definitions based on experience. A novel aspect of this approach is its ability to represent, in a manageable way, behaviour patterns involving many agents, as first-class entities that may be plugged in according to circumstances at run time. The principles are illustrated by applying them to an agent-based multimedia communications example.

-- GunterMussbacher - 13 Oct 2005

Discussion

  • Please feel free to discuss this article directly on this page. Constructive comments are welcomed! Please sign your TWiki name.

FormForVirtualLibrary edit

Title Understanding and Defining the Behaviour of Systems of Agents with Use Case Maps
Authors R.J.A. Buhr, M. Elammari, T. Gray, S. Mankovski, and D. Pinard
Type Other
Conference/Journal Title Second Conference on Practical Application of Intelligent Agents and Multi-Agent Technology (PAAM'97)
Volume/Number
Editors
Publisher
Month April
Year 1997
Pages 15
Keywords Use Case Maps, Agents, Distributed System Design, Self Modifying
Topic attachments
I Attachment Action Size Date Who Comment
4paam97.pdfpdf 4paam97.pdf manage 59 K 12 Oct 2005 - 20:28 GunterMussbacher PAAM97 poster session
4paam97.psps 4paam97.ps manage 194 K 12 Oct 2005 - 20:29 GunterMussbacher PAAM97 poster session
Topic revision: r3 - 09 Aug 2006, DanielAmyot
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback