Architectural Network Analysis: The Ultimate Cheat Sheet

Introduction

Architectural Network Analysis is a methodology that applies network theory and graph analytics to understand, analyze, and design architectural and urban spaces. It views buildings and cities as systems of interconnected elements, revealing patterns of spatial organization, movement, visibility, and social interaction. This approach bridges quantitative analysis with design intuition, helping architects and urban planners make evidence-based decisions about spatial configurations and their potential effects on human behavior, social dynamics, and environmental performance.

Core Concepts & Principles

Fundamental Concepts

Key Principles

• Configurational Analysis: Spatial relationships matter more than absolute positions
• Topological Thinking: Connections and relationships define spatial experience
• Movement Economy: Spatial configuration influences movement patterns and co-presence
• Natural Movement: Configuration itself generates movement patterns independent of attractors
• Spatial Cognition: How people understand, navigate, and remember spatial layouts
• Distributed Systems: Architecture as networks of spaces rather than isolated objects
• Quantitative-Qualitative Bridge: Network metrics can predict experiential qualities

Methodologies & Processes

Space Syntax Analysis Process

1. Define Spatial System: Determine boundaries and elements of the spatial system
2. Create Representation: Convert physical space into network representation
3. Generate Graphs: Create connectivity, visibility, or agent-based models
4. Calculate Metrics: Compute network measures (integration, choice, depth)
5. Visualize Results: Generate heat maps or colored graphs showing metric values
6. Interpret Findings: Relate quantitative results to design goals and human factors
7. Inform Design: Apply insights to refine or develop spatial configurations

Types of Network Representations

Analysis Workflow in Practice

1. Gather base information (floor plans, site plans, or 3D models)
2. Clean and prepare spatial data for analysis
3. Import into network analysis software
4. Generate appropriate network representation
5. Run analysis to calculate relevant metrics
6. Export results as visualizations and data tables
7. Correlate findings with observational data
8. Document insights and design implications

Key Metrics & Measures

Connectivity Metrics

• Degree: Number of direct connections to a node
• Connectivity Value: Number of spaces directly connected to a given space
• Control Value: Measure of relative connectivity compared to neighbors
• Neighborhood Size: Number of spaces within n steps
• Clustering Coefficient: Degree to which neighbors are connected to each other

Centrality Metrics

• Closeness Centrality: Inverse of the sum of shortest paths to all other nodes
• Betweenness Centrality: Frequency a node lies on shortest paths between other nodes
• Eigenvector Centrality: Measure of node importance based on connections to other important nodes
• Page Rank: Probability of randomly arriving at a node through network traversal
• Katz Centrality: Influence based on both immediate and distant connections

Integration Metrics

• Global Integration: Accessibility of each space from all other spaces in system
• Local Integration: Accessibility limited to spaces n steps away
• Integration [HH]: Hillier & Hanson normalized integration accounting for system size
• Mean Depth: Average number of steps from space to all other spaces
• Relative Asymmetry: Normalized mean depth value between 0 and 1

Visual Analysis Metrics

• Visual Integration: How visually connected a location is to all others
• Visual Control: Visual dominance of a location over adjacent locations
• Visual Clustering: Degree to which visually connected locations are connected to each other
• Isovist Area: Total visible area from a vantage point
• Isovist Perimeter: Boundary length of visible area
• Isovist Drift: Vector between viewpoint and center of mass of isovist
• Revelation: Rate at which new space becomes visible with movement

Software Tools & Technologies

Dedicated Network Analysis Tools

• depthmapX: Open-source platform for space syntax analysis
• SPOT (Spatial Pattern and Organization Toolkit): Integration with Rhino
• Syntax2D: Simplified space syntax plugin for AutoCAD
• Rhino Space Syntax: Network analysis tools for Rhinoceros
• QGIS Space Syntax Toolkit: GIS-based network analysis
• DeCodingSpaces: Grasshopper components for urban analysis

General Network Analysis Software

• Gephi: Open-source network visualization and analysis
• NodeXL: Excel-based network analysis tool
• NetworkX: Python library for complex networks
• igraph: Network analysis package for R and Python
• UCINET: Comprehensive social network analysis software
• Cytoscape: Open-source platform for complex network analysis

Parametric Design Integration

• Grasshopper Space Syntax components: Network analysis in parametric workflows
• LadybugTools Network Analysis: Environmental-spatial network integration
• Syntactic: Grasshopper plugin for quick space syntax analysis
• Spiderweb: Network visualization in Grasshopper
• Configurbanist: Urban configuration analysis for Grasshopper
• SmartSpaces: AI and network analysis integration

Application Domains

Building Scale Applications

Urban Scale Applications

Advanced Topics

Multi-layer Network Analysis

• 3D Spatial Networks: Extending analysis to vertical dimension
• Multi-modal Transportation: Analyzing interconnected transport networks
• Indoor-Outdoor Connection: Linking building interiors to urban contexts
• Time-based Networks: Dynamic networks changing over daily/weekly cycles
• Social-Spatial Networks: Combining social relationships with spatial proximity
• Function-Flow Integration: Relating programmatic networks to movement networks

Data Integration Methods

• GIS Data Integration: Combining spatial networks with geographic information
• Demographic Data Overlay: Correlating network metrics with population characteristics
• Sensor Data Fusion: Incorporating IoT and environmental sensing with network models
• BIM to Network Conversion: Extracting networks from building information models
• Point Cloud Processing: Generating network models from LIDAR or photogrammetry
• Real-time Data Streams: Updating network models with live occupancy or traffic data

Machine Learning Applications

• Pattern Recognition: Identifying recurring spatial configurations
• Movement Prediction: Using ML to enhance agent-based models
• Classification Models: Categorizing spaces based on network properties
• Generative Networks: Creating new spatial configurations based on learned patterns
• Optimization Algorithms: Finding optimal spatial arrangements for specific criteria
• Anomaly Detection: Identifying unusual or problematic spatial configurations

Common Challenges & Solutions

Representation Challenges

• Challenge: Choosing appropriate network representation for analysis goals
• Solutions:
  • Test multiple representation types and compare results
  • Combine complementary representations (e.g., axial and visibility)
  • Use hierarchical representations for multi-scale analysis
  • Define clear boundaries and edge conditions
  • Calibrate grid resolution for visibility analysis

Interpretation Challenges

• Challenge: Connecting quantitative metrics to qualitative experiences
• Solutions:
  • Correlate network measures with observational studies
  • Use mixed-methods approaches (surveys, tracking, interviews)
  • Develop benchmarks from successful precedents
  • Create visualization techniques that communicate to non-specialists
  • Document both metrics and experiential qualities

Technical Challenges

• Challenge: Managing computational complexity with large networks
• Solutions:
  • Simplify networks through strategic reduction
  • Use cloud computing for intensive calculations
  • Apply graph sampling techniques for large urban areas
  • Implement progressive loading of network data
  • Optimize algorithms for specific architectural applications

Best Practices & Practical Tips

Analysis Setup

• Start with clear research questions or design objectives
• Use consistent naming conventions for network elements
• Document parameters and settings used in analysis
• Maintain original data alongside processed networks
• Create baseline comparisons from similar typologies
• Verify network representation accuracy before analysis
• Calibrate models with empirical observations when possible

Results Interpretation

• Normalize metrics for comparative analysis across different systems
• Look for patterns and outliers, not just individual values
• Consider both local and global network properties
• Recognize limits of predictive capabilities
• Compare alternative designs using consistent metrics
• Combine quantitative results with qualitative assessment
• Communicate findings through clear visualizations

Design Application

• Use network analysis iteratively throughout design process
• Establish target metrics based on precedent studies
• Test design modifications with rapid analysis cycles
• Consider network resilience and redundancy
• Balance optimized metrics with experiential qualities
• Document network-based design decisions for post-occupancy evaluation
• Create feedback loops between analysis and implementation

Resources for Further Learning

Academic Programs

• UCL Bartlett Space Syntax Laboratory
• TU Delft Urban Network Analysis
• MIT City Form Lab
• Georgia Tech Shape Computation Lab
• ETH Zurich Information Architecture
• University of Michigan Taubman Complex Systems

Professional Organizations

• Space Syntax Network
• International Space Syntax Symposium
• Network Science Society
• Association for Computer-Aided Architectural Design Research in Asia (CAADRIA)
• Environmental Design Research Association (EDRA)
• Urban Complexity Network

Key Publications

• Space is the Machine by Bill Hillier
• The Social Logic of Space by Hillier & Hanson
• Decoding the City by Bernhard Streich
• Cities as Networks by Nikos Katsikis
• The Hidden Dimension by Edward T. Hall
• Network Science by Albert-László Barabási
• Space Syntax & Spatial Cognition by Christoph Hölscher

Online Learning Resources

• Space Syntax Online Training Platform
• MIT OpenCourseWare “Urban Networks”
• Coursera “Social Network Analysis”
• edX “Cities as Complex Systems”
• YouTube Space Syntax Channel
• DataCamp Network Analysis courses

Software Tutorials

• depthmapX User Manual
• Grasshopper Space Syntax Tutorials
• NetworkX Documentation
• Gephi Visualization Handbook
• QGIS Space Syntax Toolkit Guide
• DeCodingSpaces Workshop Materials

Research Centers

• Space Syntax Laboratory at UCL
• MIT Senseable City Lab
• Urban Morphology and Complex Systems Institute
• Centre for Advanced Spatial Analysis (CASA)
• ETH Zurich Future Cities Laboratory
• City Form Lab at Harvard GSD