Automated experiential engineering knowledge acquisition through Q&A contextualization and transformation

Abstract

Experiential knowledge (EK) in the brain of proficient engineers is an important asset for manufacturing enterprises. As a kind of tacit knowledge, EK is hard to describe clearly and often requires a lot of human efforts to be acquired in a computer-operable form. In this paper we propose a context-aware mechanism to acquire EK in an automatic and timely manner. The proposal comprises a formal description of EK using ontology and default logic, a machine learning-based method that discovers Q&A from the context of collaborative engineering tasks, and a semantic mapping step transforming the discovered Q&A into ontological concepts and relations. An application case shows that the EK of a group of engineers collaborating over a finite element analysis task can be automatically captured from their desktop information flow. The effectiveness of the proposed method with respect to other knowledge acquisition approaches is demonstrated through quantitative and qualitative comparison.