A QSQL-based Efficient Planning Algorithm for Fully-automated Service Composition in Dynamic Service Environments

Web service composition is emerging as a promising technology for supporting large-scale, sophisticated business process integration in a variety of complex e-science or e-business domains. Particularly, semantics have been proposed as a key to automatically solving the discovery and composition problem. However, most of semantic composition approaches still remain at a stage of low efficiency because of the performance issues brought by the involved ontology reasoning and manual processing. To address this problem, in this paper, we present a QSQL-based service composition algorithm towards a fully-automated service composition. QSQL (Quick Service Query List) is an efficient service query index list which can achieve about the same semantic service discovery effects as other existing semantic composition methods, but with much less reasoning. With our proposed QSQL-based service composition algorithm, composition plans can be created to meet a user's query in an automatic, efficient and semantic manner. In particular, with our algorithm, most existing composition plans in QSQL can be founded and ranked by exploiting a weighted Petri net representation; which will facilitate the execution verification. The final experiment is conducted to further demonstrate the feasibility of our proposed composition approach and its efficiency.