Integration of organizational workflows and the semantic web

Abstract: The Internet and the Web provide an environment to do business-to-business in a virtual world where distance is less of an issue. Providers can advertise their products globally and consumers from all over the world obtain access to these products. However, the heterogeneous and continually changing environment leads to several problems related to finding suitable providers that can satisfy a consumer's needs. The Semantic Web aims to alleviate these problems.By allowing software agents to communicate and understand the information published on the Web, the Semantic Web enables new ways of doing business and consuming services. Semantic Web technology will provide an environment where the comparison of different business contracts will be made in a matter of minutes, new contractors may be discovered continually and the organizations' routines may be automatically updated to reflect new forms of cooperation.Organizations, however, do not necessarily use the Semantic Web infrastructure to communicate internally. Consequently, to be able to gain new advantages from using Semantic Web technology, this new technology should be integrated into existing routines.In this thesis, we propose a model for integrating the usage of the Semantic Web into an organization's work routines. We provide a general view of the model as well as an agent-based view. The central component of the model is an sButler, a software agent that mediates between the organization and the Semantic Web. We describe an architecture for one of the important parts of the sButler, the process instance generation, and focus on its service retrieval capability. Further, we show the feasibility of our approach with a prototype implementation, and discuss an experiment.