Computational algorithms to evaluate design solutions using Space Syntax

Abstract

In the past, conventional computer-aided architectural design (CAAD) systems could not manage semantic information on building components and spaces but only graphical and geometric information. However, since the advent of Building Information Modeling (BIM), which has been used for managing semantic building information, determining spatial relationships as well as quantities and properties of building components in CAAD systems has become easier. It is necessary to make current CAAD systems capable of performing spatial analysis functions using BIM because they can easily recognize building components and spaces. Accordingly, this study aims to develop the computational algorithms to evaluate design solutions using Space Syntax during the process of computer-aided architectural designing. To extract topological information from design solutions, this study proposes algorithms to recognize building information produced in the form of Industry Foundation Classes (IFC), deduce the necessary topological information, and store the information in the form of matrices. The Space Syntax theory is employed to evaluate the solutions based on social properties of spaces in a building and examine the potential for adding a spatial analysis function into CAAD applications. The developed algorithms calculate the integration value for each space from spatial connectivity based on J-graphs. To validate the proposed algorithms, a program named J-Studio for Architectural Planning (J-SAP) was developed to evaluate design solutions easily and quickly. The validation results are as follows: (1) the topological information extracted from building information was decoded into a dimensionless representation and legible J-graph, (2) mathematical analyses for choosing a better design solution during computer-aided architectural designing were presented, and (3) the examination of the privacy level of each space in a building through Space Syntax analysis was discussed. Thus, this study demonstrates the possibility of determining the social properties and accessibility of spaces during the process of computer-aided architectural designing to meet client requirements by extracting topological information from building information model and performing Space Syntax analysis for evaluating alternatives using the information.