Dynamic Web Service Orchestration Architecture for the Re-Engineering of Business Processes
This thesis work proposes a BPMNSemAuto system that is used to accelerate the implementation of business process, BP, applications by using WSs. It takes a BP designed specifications as the input. After that, it generates a corresponding implemented BP application as the output. In addition, ontology is used for semantic representation of BP applications and WSs. An ontology reasoner is used to verify the syntax of BP applications, after they are transformed into the BPMN2.0 ontology. The BPMNSemAuto system provides four mains functionalities: BP syntax validation, atomic service selection, composition service discovery and implementation of BP applications. The proposed service selection algorithm considers functional and non-functional properties of WSs, service’s context and changing of QoS values over different timestamps. To the best of our knowledge, limited existing solutions consider the change QoS values in different timestamps. In addition, the proposed network based service composition algorithm applies the A* search algorithm. The heuristic function of the A* considers many criteria such as: QoS, change values of QoS over timestamps, network properties, number of service in the composite and similarity matching scores. In order to reduce the running time of the service composition process, an existing network community detection algorithm Infomap is reused for defining different network communities from a service composite network. In addition, a community labeling solution is proposed to name network communities.
Keywords: Business process re-engineering, Complex network, Ontology, Service composition, Service selection, Semantic Web service, QoS