r e s e a r c h
C o n s t r u c t i n g   D e s i g n   C o n c e p t s :   A Computational Approach to the Synthesis of Architectural Form
Kotsopoulos S, Ph.D. Dissertation, Massachusetts Institute of Technology, 2005


The scope of this study is limited to architectural design. The motivation is practical: My view is that the integration of computation in the architectural studio remains unsatisfactory. Computation is approached as a peripheral task related to representation, or entirely replaces the traditional studio techniques. Overall, computational and studio techniques remain segregated. Beginning from this problem, the thesis outlines a computational framework for design synthesis, aiming to achieve better integration of computation in the studio. Central in the thesis is the use of shape computation theory in devising descriptions for artifacts from design concepts. Using my architectural experience, I expose existing studio techniques, I make them explicit, and I model them with shape computational methods. The contributions are two: The integration of computational means such as shape algebras and rule schemata in design thinking, and the introduction of new computational paradigms for the studio.  The results and experiments of this study are presented in two parts. The first part outlines concepts that are used in the second part.

The first part, including chapters I, II, III, IV, shows what is the possible association between studio and computation and explains why. It also presents the relationship between proposed ideas and existing ideas.

Chapter I introduces preliminary questions and terms. The proposition that synthesis can be approached as calculation is also briefly discussed. Chapter II introduces basic notions of shape computation theory. The core idea of the dissertation, the construction of design concepts through computation is set: When a design concept is proposed, some framework of action is selected. This framework can be informal, or formal following compositional methods or principles. It can be personal, or driven by convention. But it is characterized by some degree of internal coherence. Chapter III deals with architectural description. Sketches, diagrams, and 3d models are used in the development of spatial concepts. The spatial calculations used in this process involve areas, volumes and their boundaries. Areas and volumes correspond to the “content” of rooms and spaces, in 2d and 3d respectively. Their boundaries are used to describe their “form”. Content and form are constantly interrelated in synthesis. Chapter IV presents the properties of a calculating device suitable for the construction of design concepts: the overlaying of multiple partial descriptions to produce a single description. The device allows heterogeneous fragments to be synthesized.

The second part of the thesis, namely chapters V, VI and VII shows how shape computation produces designs based on some design concept. Three applications in composition serve as paradigms.

Chapter V examines the compositional concept of the “domino house”. The project is based on the building program of a competition for low cost housing. The design process starts from the definition of a vocabulary of rooms, and a number of spatial relations, which describe how the rooms relate. Then, the possibilities of constructing designs from these are examined systematically. The search evolves from the definition of the “parts” (spatial vocabulary) to the construction of possible “wholes” (designs). Chapter VI presents the making of the plans for an office building. Starting from a specific site and building program, the designer proposes a design concept. This is gradually developed into a design with the aid of rule schemata and rules. Forms and relations are defined gradually on the basis of the design concept. The construction proceeds from a potential “whole” (design concept) to the definition of the “parts” (rooms and spaces). Chapter VII presents a computational interpretation of Steven’s Holl design concept for Simmons Hall undergraduate dormitory at MIT. The educational interest of this case study is to examine how rule schemata and rules can express the conceptual part of the design process. Further, since Simmons Hall is an implemented design, its educational importance extents to the comparison between “conceptual” and “actual” implementation. As it turns out certain decisions taken in the studio require revision at the stage of the actual implementation. The comparison between conceptual and actual is approached here as a tool of criticism.

The dissertation ends with the summary of results, the general conclusions, and suggestions for further research.