DocumentCode :
2112359
Title :
WAD Workflow System: Data-Centric Workflow System
Author :
Robinson, Andrew J. ; Rahayu, Wenny J. ; Dillon, Tharam
Author_Institution :
Dept. of Comput. Sci. & Comput. Eng., La Trobe Univ., Melbourne, VIC
fYear :
2009
fDate :
14-17 April 2009
Firstpage :
337
Lastpage :
344
Abstract :
To date, most scientific workflows center on the process-centric variety. This paper presents a novel approach to the workflow field by taking the view-point of a data-centric workflow. The key features of the workflow activity diagram (WAD) system are its record and field level views of workflow connections. It makes use of existing standards for visualization and provides automation features using existing domain ontologies. To execute the workflow, the WAD system makes use of the existing business process execution language (BPEL). The visualization system can be used in both process and data-centric workflows, thus lending itself towards a standard for workflow visualization. WAD handles multiple levels of user experience by using a multi-level drill-down approach. The user can drill-down as far as they are comfortable and allow the system to automate the remaining levels. The WAD system comes with two methods to automate the record and field level connections. The first is based on domain knowledge from the domain ontologies and the second is based of previous user experiences collected by the system. By connecting processes on a record and field level the system is able to connect processes that do not use the same data format. That is, it provides ad-hoc data format conversions. This removes the need for static data format convertors and thus increases system flexibility.
Keywords :
business data processing; ontologies (artificial intelligence); workflow management software; ad-hoc data format conversions; business process execution language; data-centric workflow system; workflow activity diagram system; workflow connections; workflow visualization; Australia; Automation; Computer science; Converters; Data engineering; Ecosystems; Joining processes; Ontologies; Software engineering; Visualization; Business Process Execution Language; Data-centric workflow design; Scientific Workflow;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Software Engineering Conference, 2009. ASWEC '09. Australian
Conference_Location :
Gold Coast, QLD
ISSN :
1530-0803
Print_ISBN :
978-0-7695-3599-9
Type :
conf
DOI :
10.1109/ASWEC.2009.26
Filename :
5076656
Link To Document :
بازگشت