Visual representation aid in engineering design process: From axiomatic design towards knowledge intelligence

Abstract

The notion of function has been a key element in most of the design methodologies. Its definition has been controversial within the research community, with issues arising when methodologies rely solely on one universal strict definition. Axiomatic design relies on functional decomposition, which has as a perquisite the correct definition of the main function and subfunctions of a design problem. The inability of design methodologies to be incorporated into CAD and the absence of a systematic way to store precious knowledge occurring during the design process itself becomes a hurdle for true automation and intelligence in design. This paper proposes a visual representation process of the design problem, which verifies itself through relational graph representations and the physical laws governing the structure, where they are given valuable semantics. Knowledge can be inferred, stored and analysed with the scope of making engineering design more intelligent and potentially more automated.