Towards a form processor: A framework for architectural form derivation and analysis using a formal language analogy

Form derivation is a major task in architectural design. However, to date, there does not exist a satisfactory explicit and systematic method to identify and describe this process. This dissertation constructs a theoretical and a computational framework for a form processor that helps derive and analyze forms explicitly and systematically. Moreover, it develops a database for architectural styles that helps document and classify architectonic components.; The theoretical foundation of the proposed framework consists of three major parts. These are the concepts of formal languages, Mitchell's model of artifact grammar, and the theory of reduction.; The research has four aims. The first is to demonstrate one way in which the process of form-making can be rationalized, and can, therefore, be analyzed. The second is to introduce a new direction for architectural software. This direction is based on using rudimentary architectural elements rather than basic geometric elements. The third is to develop a formal language for architectural composition. The fourth is to widen the scope, to simplify the structure of the previous related studies, and, therefore, to translate their proposals into an applicable framework that can be used by architects to develop and analyze their designs.; In terms of the methodology of the research, the theoretical design and simulation are used for theory building. The computational method and case studies are employed for theory testing.; This research is expected to have important pedagogical, theoretical, and practical consequences. It is expected to help educators and students experiment with an explicit and replicable method of deriving architectural compositions. It is also expected to enhance the acquisition of knowledge about architectural form derivation and analysis. In this sense, the research contributes to the body of testable theories in architectural research. In architectural practice, the research is expected to facilitate the creation and use of a database of rudimentary architectural elements. Such elements are often used as starting points for designing buildings. The research will also demonstrate a method for developing sophisticated ideas in simpler and faster ways than manual and traditional computer-aided methods.