An ontology‐based approach towards comprehensive workflow modelling. Issue 2 (1st April 2014)

• Record Type: Journal Article
• Authors: Koukovini, Maria N.; Papagiannakopoulou, Eugenia I.; Lioudakis, Georgios V.; Dellas, Nikolaos; Kaklamani, Dimitra I.; Venieris, Iakovos S.
• Is Part Of: IET software. Volume 8:Issue 2(2014)
• Page Start: 73
• Page End: 85
• Publication Date: 2014-04-01
• Subjects: data flow computing -- ontologies (artificial intelligence) -- workflow management software; ontology-based approach -- comprehensive workflow modelling -- core workflow perspectives -- flexibility by design principle -- formal semantics -- ontological information model -- control flow -- data flow -- in-design expression -- sophisticated security constraints; Computer software -- Periodicals; Software engineering -- Periodicals; 005.1
• DOI: 10.1049/iet-sen.2013.0037
• Languages: English
• ISSNs: 1751-8806

Abstract

Abstract : In recent years, several mature workflow modelling technologies have emerged. Nevertheless, they all present certain expressiveness limitations concerning primarily two aspects: first, none of them manages to adequately capture all three core workflow perspectives, that is, control, data and resource; further, they typically are either data‐centric or control flow oriented, being, therefore incapable of supporting domains that involve both execution patterns. In light of these issues, this study describes an innovative, highly expressive framework for workflow modelling, guided by the flexibility by design principle. Main characteristics of the proposed approach include the following: (i) it enables the comprehensive specification of workflow elements, providing extensive coverage of all aforementioned perspectives; (ii) it introduces the novel concept of assets, as a means for representing the entities being subject to the execution of workflow tasks; (iii) workflows are defined as ontologies; this, apart from the inherent benefits regarding formal semantics, offers also the advantage of their direct and transparent integration with an ontological information model; (iv) it allows the explicit modelling of both control and data flows, thus being suitable for applications based on either of them or both of them combined; and (v) its expressiveness provides for the in‐design expression of sophisticated security constraints.