Intelligent Designer: a computational approach to automate building design

The idea of using a computational approach to design our buildings is not new. Automation design projects are still developing and are far from maturity. Building design automation is even more elusive because the design process does not occur in isolation; the interaction between client (or building end user) and design has a significant impact on the final solution. 
On the other side, computers are not smart enough yet to understand clients or end usersʹ (e.g. building occupants) needs or expectations and innovate a building design, although they can offer various possibilities of assistant to the designer. However, various learning techniques provided in machine learning and artificial intelligence, such as dynamic programming and reinforcement learning, are making computers smarter at so many of the tasks humans wish to pursue. We are conducting the first stage of a broader research plan that aims to create an artificial intelligence that can make design decisions and thus alleviate architects’ and engineer designers’ effort . Based on a literature review, next section presents the new vision of building design automation in comparison to current trends in the Architecture/Engineering/Construction (AEC) industry.

Somayeh Asadi is an Assistant Professor in the Department of Architectural Engineering at the Pennsylvania State University and holds the Hartz Family Career Development Assistant Professor. Prior to that she was an Assistant Professor at the Texas A&M University. Her research interest cuts across the following themes: Automated Design, Critical Infrastructure Systems, Food-Water-Energy Nexus, Design of High Performance Buildings, and Environmental sustainability. Dr. Asadi has had several NSF, DOE, DEP, and Qatar national research foundation awards and has more than 70 journal and conference publications.