Comparison of knowledge representation in PDM and by semantic networks

Abstract

'Nowadays, computer-aided tools have enabled the creation of electronic design documents on an unprecedented scale, while determining and finding what can be reused for a new design is like searching for a 'needle in a haystack'. (…) The availability of such extensive knowledge resources is creating new challenges as well as opportunities for research on how to retrieve and reuse the knowl-edge from existing designs.' [1] If the requested knowledge is implicit (which means that it is only in the minds of the employees of a company) the retrieval and reuse of knowledge is even more com-plicated. By representing the (engineering) data backbone of a company, PDM systems are the software implementation which should support the designer to retrieve information about existing and successful design projects. This paper shows that the known data classification approaches of common PDM systems are not applicable to represent implicit (tacit) knowledge. Furthermore a new approach to knowledge representation is introduced by using Semantic Networks. The feasibility of the presented work is shown by a use-case scenario in which the conventional PDM system supported product development process is compared with the proposed way by using the soft-ware 'The Semaril' — a software tool developed at the Institute of Engineering Design/CAD based on Semantic Networks [2].