Condensing Complexity

  • K-means clustering of geometric data sets as a design tool
Scientific Publication

Authors

Daniela Kröhnert , Lukas Allner , Andrea Rossi , Hyo Wook Kim

Date

03 September 2025

Keywords

Architecture, Construction Engineering, Computer Sciences

ISBN/ISSN/ISMN, DOI

Text

This paper explores the use of K-means clustering as a design tool to manage geometric complexity in design contexts. While computational design has often been associated with formal complexity, this work investigates techniques for condensing complexity and structuring irregular datasets into meaningful clusters to enable systematic design interventions. By applying K-means clustering to irregular geometries, the paper demonstrates how digital abstraction can serve as an alternative to physical standardization, preserving material and/or geometric specificity in structured assemblies. Three case studies illustrate different approaches to integrating clustering within design systems, highlighting its potential to enable informed design decisions.

Volume/Issue, Pages

Volume 1, 297-305

Language, Format, Material, Edition

English

Activity List

Published By: Lukas Allner | Universität für Angewandte Kunst Wien | Publication Date: 10 March 2026, 14:27 | Edit Date: 16 April 2026, 15:04