Authors:
Paul Bogg
1
;
Graham Low
1
;
Brian Henderson-Sellers
2
and
Ghassan Beydoun
3
Affiliations:
1
School of Information Systems, Technology and Management, University of New South Wale, Australia
;
2
School of Software, University of Technology of Sydney, Australia
;
3
School of Information Systems and Technology, University of Wollongong, Australia
Keyword(s):
Software Development Methodologies, Method Engineering, Software Process Improvement, MOBMAS.
Related
Ontology
Subjects/Areas/Topics:
Agents
;
Artificial Intelligence
;
Artificial Intelligence and Decision Support Systems
;
Aspect-Oriented Software Development
;
Aspects
;
Distributed and Mobile Software Systems
;
Enterprise Information Systems
;
Enterprise Software Technologies
;
Knowledge Engineering and Ontology Development
;
Knowledge-Based Systems
;
Multi-Agent Systems
;
Paradigm Trends
;
Software Engineering
;
Symbolic Systems
;
Technologies for Inter-Enterprise Collaboration
Abstract:
A work product is a tangible artifact used during a software development project; for example, a requirements specifications or class model diagram. Towards a general approach for evaluating and potentially improving the quality of methodologies, this paper proposes utilizing a work product-based approach to method construction known as the “work product pool” approach to situational method engineering to accomplish this quality improvement. Starting from the final software application and identifying work product pre-requisites by working backwards through the methodology process, work product inter-dependencies are revealed. Using method fragments from a specific methodology (here, MOBMAS), we use this backward chaining approach to effectively recreate that methodology. Evaluation of the artificially recreated methodology allows the identification of missing and/or extraneous method elements and where process steps could be improved.