Search

AI for AEC The article surveys current methods of data integration, artificial intelligence (AI), optimization, and control and their (potential) applications in architecture, engineering and construction. The survey includes symbolic AI‐methods as well as subsymbolic AI methods, i. e., machine learning. The article presents these methods in the context of applications that provide insight into current research projects at the Cluster of Excellence "Integrative Computational Design and Construction for Architecture" (IntCDC) at the University of Stuttgart: (1) Data integration to link data silos in design and construction processes, (2) knowledge graphs to represent knowledge in multidisciplinary design processes, (3) automated planning for scheduling and distribution of construction tasks, (4) supervised learning to estimate the results of expensive building simulations such as operational energy or of the behavior of natural materials such as wood, (5) unsupervised learning to visualize optimization results, (6) reinforcement learning for building with fibers and bamboo, and (8) control for construction robotics. The article concludes that integrative computational design and construction requires the cooperation of humans, material, and machines, and that AI – instead of merely automating design and construction processes – can moderate this cooperation. [ABSTRACT FROM AUTHOR]