This book proposes a new class of computational tool, an Intelligent Genetic Design Tool (IGDT), intended for use in the area of conceptual design of architectural and civil engineering structures. As a computer design aid, the IGDT is innovative in its intelligent interaction with the designer. The IGDT allows the user to explore a solution space in a way which promotes the development of creative designs. Being based on Genetic Algorithms (GA's) the IGDT always submits a multiplicity of solutions (populations) for review by the designer and is thus less likely to cause design fixation than most other optimization techniques. The use of GA's also allows the designer greater latitude in exploring hard-to-code design criteria such as aesthetics, expression or meaning. In this way, it provides a service to the designer that is more useful than traditional design tools. For this dissertation, a specific application of an IGDT in the area of architectural engineering is developed. In order to exhibit the capabilities of the IGDT, examples are shown using different trussed systems under various design conditions. The ability of the IGDT to intelligently respond to the designer's preferences and assist in the discovery of useful truss topologies is demonstrated. Comparisons are made with other optimization tools and techniques. It is concluded that the IGDT offers a significantly different approach to computer-aided structural design which has the potential to enhance the designer's own creativity in discovering a good solution.
Peter von Bülow