Software development is (still) too costly and time-consuming, and the application / business context is often not sufficiently and uniformly understood by all stakeholders. The related knowledge is often just tacit and not explicitly usable or used for automation.
Semantic technology can address these issues. Ontologies offer knowledge represented declaratively, such as concepts and their relations, taxonomies, partonomies, etc. So, they are amenable for representing business concepts and objects. In contrast, however, business processes are inherently procedural and, therefore mostly not represented declaratively in practice. Due to this representational gap, important relations between business processes and their associated business concepts and objects are not yet exploited sufficiently.
We propose to study these relations for bridging this gap and for making use of them for automated generation of the business processes to be implemented in software, related to given business ontologies. The latter may already be available or (partially) found in the Web (e.g., LOD). Based on that, we will (partly) automate software development in niches for certain business applications. This will include development of an upper ontology for such a niche, and creation (generalisation) of generic business processes for such applications from known concrete ones. In addition, it will include the development of a software framework general enough for the given niche. It will be instantiated with specific ontological knowledge, which will be systematically used for instantiating related business processes for new business goals. This can be viewed as a new form of reuse of business knowledge and software, where both are tightly connected. It will result in ontology-based and (partly) automated software development guided by busiess models.