The paper discusses a current research that investigates if a computer aided conversation system can be created to support the human thought process in the early stages of architectural design. It argues that design conversations are an essential premise for designing, especially at the early stages, when the designer has to brainstorm ideas to generate creative conceptual solution-conjectures. The paper also argues that design knowledge is mainly dependent on a designer’s experiences. But experiential knowledge, stored in the long term memory is difficult to recall. Based on these arguments, an agent-based knowledge system, Design Thinker, is designed to allow for an efficient design conversation that triggers the experiential memory of the designer for recalling and associating the right experiences. It is also designed to enhance and add to the existing design knowledge of the designer by enabling them to view the knowledge through different perspective or domain lenses. The paper describes the conceptual structure of the knowledge base used in the prototype followed with a brief overview of the empirical study.


Software Engineering



URL: http://digitalcommons.calpoly.edu/cadrc/101