New report on parametric design space just released

Our new report, grounded in a review of the state-of-the-art in the pertinent areas, aims to map out the parametric space that artificially intelligent algorithms can explore, as well as mathematical formulas for measuring constraint satisfaction, efficiency, and diversity in the different design disciplines contributing to PrismArch.

Evolution at two levels of representation for dungeons. Left: the dungeon segments’ properties are represented as characters (e.g. E: empty segment) while the presence of walls (#) determine the connectivity of segments (magenta connections in middle figure). Right: each segment is evolved separately based on specifications from the left figure, and stitched together to create a complex, high-resolution dungeon.

The report surveys how current approaches in AI for architecture define representations, functional objectives and constraints, and identifies formulas, code libraries, and software that can be used for the purpose of AI-assisted content creation in PrismArch.

This document takes input from previous analysis of the user and functional requirements of PrismArch as described in deliverable 1.1 (“Report on current limitations of AEC software tools, leading to user and functional requirements of PrismArch”).

Case studies showcasing the use of topology optimization for large-scale problems related to architectural design. Left: Topology optimization for design of the upper “beam” spanning several towers for the Zendai competition. Right: Illustration for the concept design of a 288m tall high-rise in Australia, which shows the engineering and architecture expressed together: (a) problem statement, (b) results of the topology optimization, (c) renderings of *the design.*

You can download and read the full report here, under Deliverable 2.1 – Initial version of parametric design space.