Browsing by Author "Gokmen,S."

Now showing 1 - 3 of 3

Citation - Scopus: 2
Computation and Optimization of Structural Leaf Venation Patterns for Digital Fabrication
(Elsevier Ltd, 2022) Gökmen, Sabri
The morphogenetic design process of networking patterns produces anisotropic structural systems that can offer generative solutions for custom design applications. As an example of this type of pattern application, the leaf venation algorithm is introduced that can be customized through parametric inputs and density maps. This method is extended onto mesh surfaces incorporating multiple software applications combining aspects of parametric design, optimization and digital fabrication. The dynamic workflow is presented using a case study project titled “Calyx,” a public artwork completed using the computational tools developed as part of the research. The networking structural pattern of the sculpture yielded to the development of a geometry optimization process that allowed the digital fabrication of planarized structural members. The technical aspects of the design development and post-rationalization process for the construction of leaf venations patterns are discussed. © 2021 Elsevier Ltd
Citation - Scopus: 10
Computational Modeling and Analysis of Seljukid Muqarnas in Kayseri
(Association for Computing Machinery, 2022) Gökmen, Sabri; Baslk,A.; Aykln,Y.; Alacam,S.
As a historical and ornamental building element, muqarnas are widely found among the entrances of madrasas, mosques, and hans in Anatolian Seljuk architecture. In Kayseri (Turkey), muqarnas structures are characterized by symmetrical distribution of patterned geometric layers that presents computational rules for the design and construction of these ornamental structures. The presented research focuses on 12 unique muqarnas structures that are analyzed through a computational methodology combining photogrammetry, three-dimensional modeling, symmetry, and graph theory. The computational analysis shows that Seljukid muqarnas exhibit patterned branching of the symmetry axis between layers radiating from their geometric center. Using the modeled samples, the article analyzes inherent symmetry rules and growth patterns while offering a novel way of studying, modeling, and categorizing muqarnas. © 2022 Association for Computing Machinery.
Citation - Scopus: 0
Exploring Homeomorphism in Building Plans
(Association for Computing Machinery, Inc, 2020) Gokmen,S.; Gökmen, Sabri
This paper discusses a type of graph called “homeomorphically irreducible tree” (HIT) and its applicability for a formal study of symmetry in building plans. As a theoretical introduction, the mathematical properties of HITs are introduced through different historical building samples all of which display symmetry, proportion and homologous wings in their formal organization. The extracted principles are used to formulate a generative algorithm that reduces graph complexity to simple sequential numeric representation. This method is converted to a “homeomorphic machine” that is explored through generative plans. The aim of the paper is to introduce a new graph-based approach for potential morphological research into architectural symmetry. © 2020 Society for Modeling & Simulation International (SCS) centered on the notion of symmetry and branching morphogenesis in architecture. The aim of the research is two-fold. Firstly, formal computation is considered within a historical continuum where it can be applied to a broader class of historical works of architecture, potentially drawing links between them. This also requires a re-evaluation of core architectural principles such as symmetry and proportion that can potentially remedy architecture’s relationship with natural sciences. Secondly, computational methods need to be formulated primarily according to architectural principles that can overlap with various mathematical and algorithmic applications while conforming to the historical development of architectural knowledge. This aspect can have both theoretical and practical implications to architectural research and influence the formulation of new methods for machine learning in the future.