TY - GEN
T1 - Flow2Form Flow-Driven Computational Framework for Early Stage Architectural Design
AU - Kim, Frederick Chando
AU - Johanes, Mikhael
AU - Huang, Jeffrey
N1 - Publisher Copyright:
© ACADIA 2023. All rights reserved.
PY - 2023
Y1 - 2023
N2 - Flows have been a persistent theme as a rational and formal basis for architecture. This paper introduces a flow-based design framework for architecture using parametric modeling and machine learning analysis. It explores the integration of flows’ rational and figurative aspects into the early stages of the design process. The research employs parametric tools and machine learning algorithms to represent and analyze flows, focusing on the artisanal and craft processes aiming for circular proto-typology as a transformative architecture. The framework involves three stages: 3D flow modeling, machine learning analysis of formal and topological properties, and process-based programming and optimization. The results include volumetric representations of 16 artisanal flows and the classification of nodes based on their formal and topological characteristics. The framework enables the exploration of flow-driven architectural design, and bridges the gap between human interpretation and computational design. The research contributes to understanding flows to form in architecture, and the potential of machine learning in shaping architectural space.
AB - Flows have been a persistent theme as a rational and formal basis for architecture. This paper introduces a flow-based design framework for architecture using parametric modeling and machine learning analysis. It explores the integration of flows’ rational and figurative aspects into the early stages of the design process. The research employs parametric tools and machine learning algorithms to represent and analyze flows, focusing on the artisanal and craft processes aiming for circular proto-typology as a transformative architecture. The framework involves three stages: 3D flow modeling, machine learning analysis of formal and topological properties, and process-based programming and optimization. The results include volumetric representations of 16 artisanal flows and the classification of nodes based on their formal and topological characteristics. The framework enables the exploration of flow-driven architectural design, and bridges the gap between human interpretation and computational design. The research contributes to understanding flows to form in architecture, and the potential of machine learning in shaping architectural space.
UR - http://www.scopus.com/inward/record.url?scp=85192848088&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85192848088
T3 - Habits of the Anthropocene: Scarcity and Abundance in a Post-Material Economy - Proceedings of the 43rd Annual Conference of the Association for Computer Aided Design in Architecture, ACADIA 2023
SP - 408
EP - 419
BT - Proceedings Book One
A2 - Crawford, Assia
A2 - Diniz, Nancy Morgado
A2 - Beckett, Richard
A2 - Vanucchi, Jamie
A2 - Swackhamer, Marc
PB - Association for Computer Aided Design in Architecture
T2 - 43rd Annual Conference of the Association for Computer Aided Design in Architecture: Habits of the Anthropocene: Scarcity and Abundance in a Post-Material Economy, ACADIA 2023
Y2 - 21 October 2023 through 28 October 2023
ER -