Enterprise Ecosystem: Extending and Integrating Technology Serving the Enterprise

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 4 of 4
  • Item
  • Item
    A Model-driven Method to Design SoaML Services from BPMN Models: Principles, Proof-of-concept, and Validation
    ( 2023-01-03) Blal, Redouane ; Leshob, Abderrahmane ; Benzarti, Imen ; Mili, Hafedh ; Hussain, Omar K
    Today's business processes are increasingly complex as they cross organizational boundaries. To execute their business processes, organizations develop software applications called Process-Aware Information System (PAIS). PAIS designers must consider complex scenarios involving multiple partners. Consequently, the architectural design of high quality PAIS is complex and requires vast amounts of knowledge and skills both in software architecture and in the business domain. This paper proposes a model-driven method to design the architecture of PAIS using the service-oriented architecture (SOA) style. The proposed method generates SOA-based design models expressed in SoaML from the specifications of collaborative business processes expressed in BPMN. We developed a prototype tool using the Eclipse Modeling Framework (EMF) ecosystem. We tested the method on a set of processes from the Enterprise Resource Planning literature to assess its effectiveness. Our results show that 80.95\% of the identified services were relevant and corresponded to what architecture specialists expected.
  • Item
    A Total Cost of Ownership Model for Cloud Computing Infrastructure
    ( 2023-01-03) Heinrich, Simon ; Schuster, Thomas ; Kreft, Nils ; Volz, Raphael
    A holistic cost assessment of cloud computing ar-chitectures is currently hampered by the lack of assessment methods and the absence of a standardized and comprehensive total cost model. This creates uncertainty about cost developments of concrete scenarios and architectural changes. This article proposes a total cost of ownership model for cloud computing, covering the cost of adoption, procurement, migration, operation, usage, and exit. We evaluated our model in multiple application scenarios and against other models. Our model has shown to be substantially more comprehensive and applicable than other available models for cloud computing. Thus, our model can be useful both in practice and in research. We will demonstrate that our model can increase cost transparency and improve decision support.
  • Item
    On Deriving Business Models from Process Models: An Empirical Study
    ( 2023-01-03) Da Silva Torres, Isaac ; Pérez Blanco, Francisco Javier ; Fantinato, Marcelo ; Vara, Juan Manuel ; Gordijn, Jaap
    At least two business requirement perspectives of the digital ecosystem should be revisited in case such an ecosystem changes significantly: (1) the business process perspective (e.g. represented by BPMN 2.0 model), and (2) the business value perspective (e.g. depicted by an e3 value model). Although both perspectives differ largely and address different stakeholder concerns, there is also overlap between the two points of view. Moreover, often there is already an explicit understanding of how the parties in the ecosystem execute processes. However, the business value model is, in many cases, left implicit, whereas most disruptive technologies result in significant changes in the business value model. These are important to understand and analyze using a model-based approach such as the e3 value. To speed up the elicitation process, it would be more efficient to elicit an e3 value model using an already existing process model. We test a series of guidelines to derive ane3valuemodel from a given BPMN model. We conducted a controlled experiment through which we analyze the quality of a conceptual model – e3 value business value modeling (e3 value) – derived from another conceptual model – Business Process Modeling Notation v.2.0(BPMN). We measure model quality via validity and completeness with respect to a normative standard solution and an expert solution. The subjects are separated into two groups: the treatment group, which uses the guidelines to derive one model (e3 value) from the other (BPMN), and, the control group, which does not use the guidelines. Furthermore, we analyze and evaluate the implications of the experiment's results to understand the limitations and to improve them in future research.