Abstract

The main purpose of the STEP standard is to make possible the integration of product knowledge within the whole enterprise. Under this perspective, the mere exchange of geometric data is not enough, and qualitative knowledge of different kinds needs to be acquired and represented. Here, however, serious semantic problems arise, since the interpretation of the modelling primitives proposed by the standard heavily relies on implicit background knowledge. This problem has been recently underlined in [Metzger 1996], where it is argued that this background knowledge is stable enough and well agreed-upon only in the case of low-level geometric concepts. In the case of more abstract geometric concepts like design features, or non-geometric concepts like part or action, their meaning is not clear enough to be effectively shared by different application protocols. As a result, different interpretations are assumed for the same term in differ

Citations