M. J. Clayton, R. Fruchter, H. Krawinkler, J. Kunz, and P. Teicholz
We discuss the special demands for representation and reasoning posed by crossdisciplinary communication in collaborative conceptual building design. These needs are not well addressed by existing software technology. By building upon design theory literature and observations of a case study of an actual building design project, we are identifying and devising computational strategies for addressing these needs. Our conceptualization suggests that designers propose a shared form model, interpret the form model into discipline models, critique the discipline form models to derive behavior and compare it to function, and explain the results to other members of the team. We present this propose-interpret-critique-explain paradigm as a communication cycle for collaborative conceptual building design. We are exploring and testing the conceptualization by modeling it with an experimental software prototype, ICM, that integrates graphic representations of, and AI reasoning about, the evolving building design.