An Overview of Knowledge Engineering and its Relevance to CAAD

TLDR: Developments in artificial intelligence have opened up ways of increasing the applicability of computers by acquiring and representing knowledge in computable forms, and these approaches supplement rather than supplant existing uses of computers.

Abstract: Computer-aided architectural design (CAAD) has come to mean a number of often disparate activities. These can be placed into one of two categories: using the computer as a drafting and, to a lesser extent, modelling system; and using it as a design medium. The distinction between the two categories is often blurred. Using the computer as a drafting and modelling tool relies on computing notions concerned with representing objects and structures numerically and with ideas of computer programs as procedural algorithms. Similar notions underly the use of computers as a design medium. We shall return to these later. Clearly, all computer programs contain knowledge, whether methodological knowledge about processes or knowledge about structural relationships in models or databases. However, this knowledge is so intertwined with the procedural representation within the program that it can no longer be seen or found. Architecture is concerned with much more than numerical descriptions of buildings. It is concerned with concepts, ideas, judgement and experience. All these appear to be outside the realm of traditional computing. Yet architects discoursing use models of buildings largely unrelated to either numerical descriptions or procedural representations. They make use of knowledge about objects, events and processes - and make nonprocedural (declarative) statements that can only be described symbolically. The limits of traditional computing are the limits of traditional computer-aided design systems, namely, that it is unable directly to represent and manipulate declarative, non-algorithmic, knowledge or to perform symbolic reasoning. Developments in artificial intelligence have opened up ways of increasing the applicability of computers by acquiring and representing knowledge in computable forms. These approaches supplement rather than supplant existing uses of computers. They begin to allow the explicit representations of human knowledge. The remainder of this chapter provides a brief