Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Spatial analysis›Network-Based Spatial Analysis
Regression modelGIS / spatial

Network-Based Spatial Analysis

Also known as: network spatial analysis, network-constrained spatial analysis, spatial network analysis, NBSA

Network-based spatial analysis (NBSA) analyzes the distribution and interaction of spatial phenomena constrained to a network structure — such as roads, railways, or rivers — using network distance rather than straight-line (Euclidean) distance. It is the appropriate framework whenever movement, proximity, or risk is governed by the underlying network topology rather than open space.

ScholarGate
  1. Regression model
  2. v1
  3. 2 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Network-Based Spatial Analysis
Geographically Weighted…Hot Spot AnalysisSpatial AutocorrelationLocal Kernel Density Est…Local Network-Based Spat…Panel Network-Based Spat…Remote Sensing Classific…

When to use it

Use NBSA whenever the phenomenon of interest is constrained to a network: road traffic incidents, crime on street segments, pedestrian flows, pipeline failures, or species distribution along river networks. It is essential when Euclidean proximity is a poor proxy for actual accessibility or risk. Do not use it when events genuinely occur across an open planar surface (e.g., spread of an airborne pollutant across a field) — standard planar spatial analysis is more appropriate in that case. It also requires a reliable, topologically correct network dataset; poorly digitized networks undermine the analysis.

Strengths & limitations

Strengths
  • Measures distance and proximity in a way that matches how movement or risk actually operates along infrastructure.
  • Avoids the false positives of planar hotspot analysis where Euclidean clusters span impassable barriers.
  • Network KDE produces density estimates that respect connectivity, giving more realistic exposure or risk surfaces.
  • Directly usable for resource allocation on networks (optimal placement of emergency services, sensors, or stops).
  • Well-supported in GIS software (ArcGIS Network Analyst, QGIS, SANET toolbox, R spdep/sfnetworks).
Limitations
  • Requires a clean, topologically valid network dataset — gaps, dangling edges, or wrong directionality corrupt all downstream results.
  • Computationally intensive for large networks because all pairwise shortest paths must be computed or approximated.
  • Network KDE bandwidth selection lacks universal rules; results are sensitive to the chosen bandwidth h.
  • Cannot easily incorporate off-network phenomena (e.g., a pedestrian cutting across a park).
  • Statistical inference (significance tests) typically relies on Monte Carlo simulation, which can be slow.

Frequently asked

How is network-based spatial analysis different from standard GIS buffer analysis?

Buffer analysis draws circular zones using Euclidean distance, ignoring barriers and the actual path people or vehicles must travel. NBSA uses shortest-path network distance, so a location 200 m away by road may be closer in network terms than a location 150 m away across a river. This matters whenever movement is constrained to the network.

What software can run network-based spatial analysis?

SANET (free toolbox for ArcGIS, also available standalone), ArcGIS Network Analyst (commercial), the R packages sfnetworks and spNetwork, and the Python library OSMnx combined with NetworkX all support NBSA workflows. QGIS has basic network analysis tools via the QNEAT3 plugin.

What network data quality is needed before analysis?

The network must be topologically consistent: edges must connect at shared nodes, no dangles or gaps, correct directionality for one-way links, and projected in a planar coordinate system with meaningful length units. OpenStreetMap data often needs cleaning with tools like osmnx.simplify_graph or PostGIS topology validation before use.

How do I choose the bandwidth for network kernel density estimation?

Common choices are 100–500 m for pedestrian-scale phenomena and 500–2000 m for vehicle-scale phenomena. Cross-validation or likelihood-based methods (as implemented in the R spNetwork package) give data-driven bandwidth selection. Always report sensitivity of conclusions to bandwidth choice.

Can I apply Moran's I to network-constrained data?

Standard Moran's I uses a planar spatial weights matrix and is not appropriate for network data. A network-adapted weights matrix — where neighbors are defined by adjacency along the network and weights by inverse network distance — should be constructed instead, and significance tested by Monte Carlo permutation on the network.

Sources

  1. Okabe, A., Satoh, T., Furuta, T., Sugihara, K., & Okano, K. (2006). Generalized network Voronoi diagrams: Concepts, computational methods, and applications. International Journal of Geographical Information Science, 22(9), 965–994. DOI: 10.1080/13658810701587891 ↗
  2. Okabe, A., & Sugihara, K. (2012). Spatial Analysis Along Networks: Statistical and Computational Methods. Wiley. ISBN: 978-0470770818

How to cite this page

ScholarGate. (2026, June 3). Network-Based Spatial Analysis. ScholarGate. https://scholargate.app/en/spatial-analysis/network-based-spatial-analysis

Related methods

Geographically Weighted RegressionHot Spot AnalysisSpatial Autocorrelation

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Geographically Weighted RegressionSpatial analysis↔ compare
  • Hot Spot AnalysisSpatial analysis↔ compare
  • Spatial AutocorrelationSpatial analysis↔ compare
Compare side by side →

Referenced by

Local Kernel Density EstimationLocal Network-Based Spatial AnalysisPanel Network-Based Spatial AnalysisRemote Sensing Classification

Similar methods

Local Network-Based Spatial AnalysisSpace-Time Network-Based Spatial AnalysisNetwork Distance AnalysisService Area AnalysisStreet Network AnalysisPanel Network-Based Spatial AnalysisSpatial Design Network Analysis (sDNA)Urban Network Analysis

Related reference concepts

GIS and Spatial Analysis in ArchaeologyNetwork Analysis in the HumanitiesTransport GeographyNetwork AnalysisGeographic Information ScienceGraph and Network Visualization

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Network-Based Spatial Analysis (Network-Based Spatial Analysis). Retrieved 2026-07-21 from https://scholargate.app/en/spatial-analysis/network-based-spatial-analysis · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Atsuyuki Okabe and colleagues
Year
1990s–2000s
Type
Spatial network model
DataType
Point events or flows on a network (roads, rivers, infrastructure)
Subfamily
GIS / spatial
Related methods
Geographically Weighted RegressionHot Spot AnalysisSpatial Autocorrelation
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account