May 23-August 19, 2011
In the IDEA Studio, researchers from the University of Southern California sought to revolutionize early stage building design by “designing in” energy and economics through automation and evolutionary optimization.
Using Project Vasari, Revit Architecture, and the Green Building Studio application programming interfaces, the researchers developed a plug-in to automatically generate, analyze, visualize, and rank various design configurations according to user-defined parameters, such as energy use intensity and financial goals.
Embedding their system with a genetic algorithm, the team created a method for essentially breeding the best designs based on customized parameters. Using the system to evaluate a complex form with 2 twisting towers resulted in an expansive solution space with varying geometries.
Some designs were stronger than others in the vast solution space. The top contenders reflected the best mix of strong financial performance and design scores with minimized energy use. To reduce subjectivity in assigning design scores, a quantitative approach was devised. Values were assigned to achievement of design intent, consideration of site constraints, and variability of level-to-level height and space usage.
The result offers the potential for multidisciplinary design teams to systematically explore a large number of design options quickly to find the best alternative.