CACDA: A knowledge graph for a context-aware cognitive design assistant

The design of complex engineered systems highly relies on a laborious zigzagging between computer-aided (CAD) software and rules prescribed by manuals. Despite the emergence knowledge management techniques (ontology, expert system, text mining, etc.), companies continue to store in large unstructured documents. To facilitate integration CAD software, we propose graph that structures set computable format. organises entities (nodes), relationships among (edges), as well contextual information. categorisation four sub-contexts: semantic, social, engineering, IT – facilitates development data model, especially definition “design context” concept. paves way context-aware cognitive assistant. Indeed, connected or embedded assistant will capture context near real time run reasoning operations extend traditional capabilities, such recommendation rules, verification solutions, automation routines. Our validation experiment shows current version is more efficient than document-based design. On average, participants using an document have precision 0.36 whereas our demonstrator obtain 0.61 score. Finally, designers supported spend designing searching for applicable compared approach.