[Paper Review] Urban volumetrics: spatial complexity and wayfinding, extending space syntax to three dimensional space

Lingzhu Zhang, Alain Chiaradia|arXiv (Cornell University)|Dec 28, 2020

Urban Design and Spatial Analysis78 references4 citations

TL;DR

This paper introduces Urban Volumetrics, a novel 3D spatial analysis framework that extends space syntax to three-dimensional built environments by developing a hybrid angular-Euclidean line representation. Using a large-scale case study with 17,307 observed pedestrian movements, it demonstrates that higher-dimensional completeness in spatial representation significantly improves prediction of wayfinding behavior, validating the need for full 3D line-based modeling in complex indoor-outdoor urban spaces.

ABSTRACT

Wayfinding behavior and pedestrian movement pattern research relies on objective spatial configuration representation and analysis, such as space syntax, to quantify and control for the difficulty of wayfinding in multi-level buildings and urban built environments. However, the space syntax's representation oversimplifies multi-level vertical connections. The more recent segment and angular approaches to space syntax remain un-operationalizable in three dimensional space. The two dimensional axial-map and segment map line representations are reviewed to determine their extension to a novel three dimensional space line representation. Using an extreme case study research strategy, four representations of a large scale complex multi-level outdoor and indoor built environment are tested against observed pedestrian movement patterns N = 17,307. Association with the movement pattern increases steadily as the representation increases toward high three-dimensional space level of definition and completeness. A novel hybrid angular-Euclidean analysis was used for the objective description of three dimensional built environment complexity. The results suggest that pedestrian wayfinding and movement pattern research in a multi-level built environment should include interdependent outdoor and indoor, and use full three-dimensioanal line representation.

Motivation & Objective

• To address the limitations of traditional space syntax in representing vertical and multi-level spatial configurations in 3D urban environments.
• To develop a fully operationalizable 3D spatial representation that captures interdependent indoor and outdoor spatial complexity.
• To test whether increasing the dimensional completeness of spatial representation improves the accuracy of predicting pedestrian movement patterns.
• To validate a novel hybrid angular-Euclidean analysis for quantifying 3D built environment complexity.
• To establish a methodological foundation for future research on wayfinding and spatial cognition in complex 3D urban and architectural spaces.

Proposed method

• Adapts and extends 2D space syntax representations—axial maps and segment maps—into a three-dimensional line-based spatial representation.
• Develops a hybrid analytical approach combining angular and Euclidean metrics to describe spatial complexity in 3D space.
• Employs an extreme case study strategy using a large, complex multi-level built environment integrating indoor and outdoor spaces.
• Compares four increasingly detailed spatial representations (2D, 3D axial, 3D segment, full 3D line) against observed pedestrian movement data (N = 17,307).
• Uses statistical association analysis to evaluate how well each representation predicts actual pedestrian movement patterns.
• Applies a novel 3D spatial configuration model that maintains topological and geometric fidelity across vertical and horizontal dimensions.

Experimental results

Research questions

• RQ1 How does the inclusion of three-dimensional spatial configuration improve the prediction of pedestrian movement patterns compared to 2D representations?
• RQ2 To what extent does increasing the completeness and dimensionality of spatial representation enhance the accuracy of wayfinding behavior modeling?
• RQ3 Can a hybrid angular-Euclidean analysis effectively quantify spatial complexity in multi-level 3D built environments?
• RQ4 How do interdependent indoor and outdoor spatial configurations influence pedestrian navigation in complex urban settings?
• RQ5 Is there a measurable improvement in predictive power when using full 3D line-based representations over simplified 2D or 3D projections?

Key findings

• The association between spatial representation and observed pedestrian movement increased steadily with each higher level of 3D spatial definition, from 2D to full 3D line representation.
• The full 3D line representation demonstrated the strongest predictive power for pedestrian movement patterns, outperforming 2D and partial 3D models.
• The hybrid angular-Euclidean analysis successfully captured the complexity of 3D spatial configurations in a way that was operationally meaningful and quantitatively measurable.
• The study confirms that interdependent indoor and outdoor spatial configurations significantly influence wayfinding behavior and must be modeled together.
• The results validate that 3D spatial representation is essential for accurately modeling wayfinding in complex, multi-level urban environments.
• The methodological framework enables objective, scalable, and replicable analysis of spatial complexity and pedestrian movement in 3D urban built environments.