In this thesis, we propose a bundled solution to address the limitations of peer-to- peer (P2P) workflow management systems such as the architectural weakness and the lack to support exception handling at instance level. The aim is to conceptualise, formalise, implement and validate this bundled solution. After an in-depth analysis of the above mentioned problems, a novel decentralized P2P workflow architecture (P-FLOW) has been developed. Then, an innovative mechanism is proposed to handle the instance level exceptions in P-FLOW. This mechanism ensures the handling of exceptions in a coordinated manner and separates the exception logic from the business logic. Furthermore, we proposed a novel feasible and flexible intelligent problem solving approach which facilitates more common understanding and exception handling experience sharing in P2P WFMSs. In our approach, a case- based reasoning (CBR) based exception handling mechanism with integrated human involvement is used. The CBR enhances the exception handling capabilities by capturing and collecting experience in handling exceptions and storing them in the form of cases. Then, these cases will be retrieved and reused for new similar situations. This mechanism allows the workflow participants to handle exceptions which need to be managed in similar way, but may occur in different instances. In our work, the workflow participants describe the exceptions in flexible way using natural language and store them as cases in case repository for future use. Then natural language processing techniques such as sentence similarity measures are used to retrieve the relevant information within CBR methodology. The feasibility of the main concepts presented in this thesis has been demonstrated using a prototype implementation. Further work has been discussed to apply the proposed P-FLOW in inter-organizational workflow scenario and potential system enhancement using agent technology.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:554315 |
| Date | January 2012 |
| Creators | Aldeeb, Aibrahim Abdusamed |
| Publisher | Manchester Metropolitan University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0019 seconds