Semantic representation of engineering knowledge: pre-study

Artificial intelligence, AI, is having a strong hype in research and in industry. The main factor for the hype is the fast progress in the development of data-based AI technologies, especially machine learning in all of its forms. The technologies and solutions have enabled fast algorithms to recognise features in data or to make quick decisions based on given inputs. The common feature in these applications is that either the approach requires a large amount of data or it can analyse and e.g. classify a large amount of data. The need for increased computer intelligence also in engineering design is evident, but one narrow set of methods and technologies does not solve the challenges. The solution is to integrate different methods and technologies to fulfil the different kinds of needs. On one hand, we need fast algorithms e.g. to help the user in his/her tasks or to categorise large data sets, while on the other hand we also need to utilise the existing domain knowledge, especially in the design of complex products and systems. The previous hype of artificial intelligence in 1990’s and 2000’s was emphasising knowledge representation, management and engineering, and one of the main outcomes of the hype has been the set of technologies for the Semantic Web. In this work, we revisit the selected technologies of the Semantic Web and study the state-of-the-art applications utilising them for representing engineering design data (Sections 1–6). In addition, some of the existing tools and systems for editing data models as well as for storing the knowledge data were studied and are presented in Section 7. To illustrate the overall approach of engineering knowledge representation,
a small-scale case study was done and is presented in Section 8. A brief summary of the report is presented in Section 9.