[agents] Design Process Documentation and Fragmentation Working Group **call for proposals**

[Massimo Cossentino]

IEEE FOUNDATION FOR INTELLIGENT PHYSICAL AGENTS

STANDARDS COMMITTEE (FIPA SC)

Design Process Documentation and Fragmentation Working Group

Chair: Cossentino Massimo (cossentino at pa.icar.cnr.it)

Official Website: http://www.fipa.org/subgroups/DPDF-WG.html

Working website: http://www.pa.icar.cnr.it/cossentino/fipa-dpdf-wg/


CALL FOR PROPOSALS



Problem Statement:

Complex software system development with the agent-oriented approach  
requires suitable agent oriented modeling techniques and processes  
providing explicit support for the key abstractions of the agent  
paradigm.

Several design processes supporting analysis, design and  
implementation of Multi-Agent Systems (MAS) have been to date proposed  
in the context of Agent Oriented Software Engineering (AOSE).

Each of them presents different advantages when applied to specific  
problems. A unique design process cannot be general enough to be  
useful to everyone without some kind of customisation; when developing  
a new design process, several MAS developers/designers prefer to use  
phases or models or elements coming from existing design processes in  
order to build up a personalised approach for their own problem.

This problem can be faced by adopting the method engineering paradigm  
(more precisely the situational method engineering paradigm).  
Situational method engineering paradigm provides means for  
constructing ad-hoc software engineering processes following an  
approach based on the reuse of portions of existing design processes,  
the so called method fragments stored in a repository, called method  
base.

Method fragment or simply fragment is the building block of ad-hoc  
design processes; several well known approaches in literature present  
different definitions and descriptions of method fragment but all of  
them share the same assumption: each existing design process can be  
considered as composed of self contained components, the fragments.  
The definition of method fragment in every kind of situational method  
engineering approach constitutes the base for the extraction of  
fragments from existing design processes, for their retrieval from the  
method base and for their assembly in the new design process. Today it  
does not exist (yet) a unique, standard, definition of method fragment  
besides when looking at existing design processes, usually designers  
have to cope with an additional difficulty consisting in the lack of  
uniformity in their documentation.

We think that in the FIPA context, the adoption of a standardised way  
for representing design aimed at developing multi-agent systems is  
very advisable. This would facilitate the comparison of different  
processes, their evaluation and an easier transition to the next step  
of our proposal: the fragmentation of such processes in a set of  
fragments abiding to the same structure. The definition of a standard  
way for documenting design processes and the definition of a standard  
structure of the fragments that compose these processes are therefore  
the aims of this working group.


Objective of this working group:

This working group aims to propose a definition of method fragment to  
be used during a situational method engineering process, the  
fundamental elements it is composed of and the metamodel it is based on.

The first step will be the identification of the most suitable process  
metamodel and notation: (i) for the representation of the existing  
design processes from which the fragments have to be extracted, and  
(ii) for the representation of fragments themselves. An important  
contribution about that might come from an OMG specification, the  
Software Process Engineering Metamodel 2.0 (SPEM), or from the ISO  
24744 specification.

The second step will consist in the definition of a proper template  
for the description of agent-oriented design processes. Such a  
template will, obviously, refer to the selected process metamodel and  
suggest the adoption of good practices in documenting existing  
processes as well as defining new ones.

The third step will be the definition of the Method Fragment Structure  
and Documentation Template. This work will start from the results  
obtained by the FIPA Methodology Technical Committee (years 2003-2005).

The documentation and fragmentation of some existing design processes  
(members already committed for: ADELFE, ASEME, ASPECS, INGENIAS,  
GORMAS, PASSI, SODA) will be performed with the aim of defining  
guidelines and best practices for process fragmentation according to  
the specification provided in the Method Fragment Structure and  
Documentation Template.


The proposed work aims at providing the possibility of representing  
design processes and method fragments through the use of a  
standardized structure thus allowing the creation of sharable  
repositories and enabling an easier composition of new design processes.

In the future this standardized way of representing method fragments  
could be used for the implementation of fragments in a CAPE (Computer  
Aided Process Engineering) tool supporting the designer during the  
construction phases.




Objective of this call for proposal:

The first step of this WG activity consists in identifying the proper  
approach to the documentation of agent-oriented design processes.This  
approach should include:

•   A description of a design process metamodel.

•   A notation (coherent with the above listed design process  
metamodel).

•   A template of document to be used for describing existing design  
processes

The second step is devoted to the adoption of a method fragment  
metamodel (obviously coherent with adopted process description  
approach) and the description of process fragmentation guidelines/ 
procedures.


With this call for proposal, WG members ask for the contribution of  
scientists, practitioners and experts in the area in order to receive  
proposals suitable for satisfying the above described requirements.


How to contribute:

Initial proposal can also be partial or they can even address only a  
portion of the overall problem. It is in the policy of this group to  
discuss and consider all contributions.

Contributions can be submitted by using the WG mailing list (if the  
submitter is already subscribed to that) or they can be submitted  
contacting the chair.


Deadline:

Although new interesting ideas are always welcome, it has to be  
considered that starting from the beginning of July 2009, this working  
group will deal with the definition of the process metamodel and the  
specification of the related notation.




List of members (in adhesion chronological order):

Submitters:

•       M. Cossentino (National Research Council, IT),

•       A. Molesini (Univ. Bologna, IT)

•       A. Omicini (Univ. Bologna, IT)

•       V. Seidita (University of Palermo, IT)

•       G. Cabri (Univ. Modena e Reggio Emilia, IT)

•       M.P. Gleizes (Univ. Paul Sabatier, FR)

•       V. Hilaire (Univ. Belfort, UTBM, FR)

•       J. Pavon (Univ. Madrid, ES)

•       Frédéric Migeon (Univ. Paul Sabatier, FR)

•       Carole Bernon (Univ. Paul Sabatier, FR)

•       Juan Carlos Gonzalez Moreno (Univ. de Vigo, ES)

•       Rubén Fuentes Fernandez (Univ. Madrid, ES)

•       Piero Storniolo (National Research Council, IT)


Supporters:

•       S. DeLoach (Kansas State Univ., USA)

•       G. Fortino (Univ. Calabria, IT)

•       S. Galland (Univ. Belfort, UTBM, FR)

•       A. Garro (Univ. Calabria, IT)

•       N. Gaud (Univ. Belfort, UTBM, FR)

•       Z. Guessom (Univ. LIP6, Paris, FR)

•       M.P. Huget (Univ. Savoie, FR)

•       V. Morreale (Engineering Ingegneria Informatica, IT)

•       S. Ossowski (Univ. Rey Juan Carlos, ES)

•       J. Pena (Univ. Seville, ES)

•       W. Renz (Hamburg University of Applied Sciences, DE)

•       J. Sudeikat (Hamburg University of Applied Sciences, DE)

•       A. Gomez Rodriguez(Univ. de Vigo, ES)