Designer Centered Development of GA-Based DSS for Conceptual Design of Buildings

Abstract

Research into computer-assisted design is often undertaken with relatively little input from practicing designers. Without such involvement, one can argue that the techniques developed are not subjected to any form of evaluation or independent scrutiny and thus the methodology is wanting in terms of scientific rigor. The paper describes a research project to examine the use of evolutionary computing techniques for the design of commercial office-type buildings. The research has involved a considerable amount of input from designers, and the contributions that they make to the research are identified and discussed. In particular, the benefits gained from their evaluation of the software are considered. Also, the possible pitfalls regarding such an approach to research are identified.