OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks

Geoff Boeing

article2016Computers, Environment and Urban Systems1,660 citations

Introduces OSMnx, a Python package that automates the extraction, topological correction, and multi-scale spatial analysis of complex street networks directly from OpenStreetMap data.

Listen

Urban planning and transportation research frequently face major methodological bottlenecks when analyzing street networks. Existing empirical studies often rely on small sample sizes of 10 to 50 networks due to the high friction of data collection, oversimplify real-world systems into flat two-dimensional graphs that misrepresent infrastructure such as bridges and tunnels, and suffer from poor replicability caused by ad hoc definitions. These constraints limit the ability of decision-makers and planners to evaluate network performance, resilience, and connectivity consistently at scale.

The article introduces and evaluates OSMnx, a free, open-source Python software package designed to automate the collection, topological correction, visualization, and spatial analysis of complex street networks anywhere in the world using OpenStreetMap data.

OSMnx enables researchers to construct topologically accurate, non-planar directed graphs natively with minimal code. To establish and demonstrate this capability, the article outlines the tool's core architecture—combining network analysis, geographic information systems, and elevation data—and applies it to a comparative case study across three 0.5-square-kilometer neighborhood sections in Portland, Oregon: Downtown, Laurelhurst, and Northwest Heights.

The analysis demonstrates several key findings regarding urban form, connectivity, and network vulnerability. In terms of density and scale, Downtown exhibits a fine-grained, dense layout with 164 intersections per square kilometer and an average segment length of 76 meters, compared to Northwest Heights' coarse-grained layout of 28 intersections per square kilometer and an average segment length of 117 meters. Topological analysis revealed that Downtown's network initially appeared less resilient to disruption, requiring only 1.3 node failures on average to disconnect paths between random locations, compared to 2.1 in Laurelhurst; this reduced resilience stems entirely from Downtown's strict one-way street network. When modeled as two-way streets, Downtown's resilience more than doubled to 2.9 paths, making it the most robust of the three. Additionally, betweenness centrality analysis revealed significant bottleneck risks: in Northwest Heights, a single critical intersection handles 43% of all shortest paths, whereas Downtown’s most central node handles only 15%, indicating high vulnerability to single-point disruptions in suburban-style layouts.

These findings have direct operational and policy implications for urban design and infrastructure management. Street directionality heavily penalizes network redundancy, showing that converting one-way corridors to two-way configurations can substantially improve traffic resilience and circulation options. Furthermore, measuring network centrality highlights critical choke points, allowing city agencies to prioritize specific intersections for safety enhancements, disaster mitigation, and maintenance before disruptions occur.

Transportation planners and researchers should leverage automated spatial tools to conduct large-scale, reproducible network evaluations rather than relying on manual, small-sample approaches. When evaluating circulation systems, analysts should account for true directionality and non-planar structures to avoid distorted metrics. Future work should expand analyses beyond isolated neighborhood subsets to prevent boundary edge effects and incorporate richer streetscape attributes as crowdsourced databases expand.

Confidence in the tool’s spatial and topological accuracy is high, though readers should note that the analysis depends on the completeness of OpenStreetMap data, which is extensive across the United States and Europe but less detailed in parts of the developing world. Additionally, the illustrative case study reflects small sample boundaries that omit surrounding regional traffic flows, meaning broader municipal conclusions should rely on large-scale network datasets.

University of California Berkeley graph theory network analysis network architecture OSMnx complex street networks urban network analysis

arXiv: 1611.01890 gboeing/osmnx

Paper: Spatial Networks, Marc Barthelemy (2010). Provides the foundational theoretical principles of spatial constraints, planarity, and centrality anomalies in geographically embedded networks like street networks.

Paper: Spatio-temporal Graph Convolutional Neural Network: A Deep Learning Framework for Traffic Forecasting, Bing Yu et al. (2017). Applies deep graph convolutional modeling to predict dynamic traffic flows over spatial road network topologies like those constructed and exported by OSMnx.
Paper: Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting, Yaguang Li et al. (2017). Extends spatial network analytics by modeling traffic dynamics as a diffusion process over directed sensor road graphs.
Paper: T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction, Ling Zhao et al. (2018). Leverages road network topology and recurrent units to forecast temporal traffic conditions on urban street layouts.
Paper: Graph WaveNet for Deep Spatial-Temporal Graph Modeling, Zonghan Wu et al. (2019). Advances deep spatio-temporal modeling on road networks by learning adaptive spatial dependencies alongside fixed network connectivity.
Paper: Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting, Shengnan Guo et al. (2019). Introduces spatial-temporal attention mechanisms on graph-structured road topologies to forecast multi-scale traffic flows.
Paper: Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting, Chao Song et al. (2020). Develops synchronous spatial-temporal graph convolutions for forecasting localized traffic patterns over real-world road networks.
Paper: Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting, Lei Bai et al. (2020). Enhances traffic prediction on urban road networks by adaptively learning node-specific parameters and hidden spatial relations.
Paper: Attention, Learn to Solve Routing Problems!, Wouter Kool et al. (2018). Applies attention-driven reinforcement learning to solve vehicle routing and traveling salesperson problems across spatial networks.
Paper: Learning Combinatorial Optimization Algorithms over Graphs, Elias Boutros Khalil et al. (2017). Combines graph neural network representations and reinforcement learning to automate combinatorial optimization heuristics on network graphs.
Paper: Road Extraction by Deep Residual U-Net, Zhengxin Zhang et al. (2017). Presents a deep learning approach for extracting road networks and spatial geometry from remote sensing imagery.

Cover for OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks

Abstract

Urban scholars have studied street networks in various ways, but there are data availability and consistency limitations to the current urban planning/street network analysis literature. To address these challenges, this article presents OSMnx, a new tool to make the collection of data and creation and analysis of street networks simple, consistent, automatable and sound from the perspectives of graph theory, transportation, and urban design. OSMnx contributes five significant capabilities for researchers and practitioners: first, the automated downloading of political boundaries and building footprints; second, the tailored and automated downloading and constructing of street network data from OpenStreetMap; third, the algorithmic correction of network topology; fourth, the ability to save street networks to disk as shapefiles, GraphML, or SVG files; and fifth, the ability to analyze street networks, including calculating routes, projecting and visualizing networks, and calculating metric and topological measures. These measures include those common in urban design and transportation studies, as well as advanced measures of the structure and topology of the network. Finally, this article presents a simple case study using OSMnx to construct and analyze street networks in Portland, Oregon.

Topics

graph theory network analysis network architecture topological structure

Keywords

OSMnx complex street networks urban network analysis OpenStreetMap streets urban street patterns spatial network network topology analytical urban design planar graphs shortest paths

1. Introduction
2. Background
2.1. Graphs and networks
2.2. Representation of street networks
2.3. Street network analysis
2.4. Current tool landscape
2.5. Research problem
3. OSMnx: Functionality and comparison to existing tools
3.1. Acquire political boundaries and building footprints
3.2. Download and construct street networks
3.3. Correct and simplify network topology
3.4. Save street networks to disk
3.5. Analyze street networks
4. Case study: Portland, Oregon
5. Discussion
6. References
Appendix: Code and Data
6.1. Code examples
6.2. Data repository

Knowls

Knowl 1 — OSMnx Graph Representation of Urban Street Networks
model/method
OSMnx models urban street networks as primal, non-planar, directed multigraphs with self-loops ( $G = (V, E)$ ). In this representation, street intersections and dead-ends are nodes ( $V$ ), and street segments connecting them are directed edges ( $E$ ).

Unlike planar graph models that split crossing street polylines at two-dimensional intersections, OSMnx preserves non-planar 3D reality, preventing false nodes from being generated where grade-separated infrastructure (such as bridges, tunnels, and overpasses) crosses other streets without an actual physical junction.

Directionality is natively preserved: one-way streets are represented as single directed edges ( $u \to v$ ), while bidirectional streets are represented by a pair of reciprocal directed edges ( $u \to v$ and $v \to u$ ). Multigraph support allows parallel edges between the same two nodes when distinct physical routes connect them. Each edge retains its true spatial polyline geometry, length (in meters), street type classifications, and optional attributes such as node elevations and segment grades.
Knowl 2 — Spatial Data Querying, Buffering, and Truncation Pipeline
algorithm
To construct street networks for specific geographical zones without introducing boundary truncation artifacts, OSMnx implements an automated spatial querying and filtering pipeline:
1. Input Query: Accepts an administrative place name (e.g., city, county, or neighborhood), a geographic boundary polygon, a bounding box, or an address/coordinate pair with a network or Euclidean distance buffer.
2. Geocoding and Geometry Retrieval: For named places, queries OpenStreetMap's Nominatim API to obtain the formal boundary polygon.
3. Perimeter Buffering: Expands the boundary polygon outward by a 500-meter buffer.
4. Data Acquisition: Queries OpenStreetMap's Overpass API to download all street polylines and nodes lying within the buffered polygon according to the requested network type (drive, drive_service, walk, bike, all, or all_private).
5. Graph Construction and Simplification: Assembles the raw data into a directed multigraph and executes topological simplification.
6. Boundary Truncation: Truncates the simplified network back to the original unbuffered polygon. Buffering prior to truncation ensures that peripheral intersections at the border are not falsely classified as dead-ends or assigned deflated node degrees due to streets continuing outside the study area boundary.
7. Coordinate Projection and Elevation: Algorithmically calculates the appropriate Universal Transverse Mercator (UTM) zone based on the geometry's centroid to project the network, and optionally queries the Google Maps Elevation API to compute node elevations and street segment grades.
Knowl 3 — Algorithmic Network Topological Simplification and Node Filtering
algorithm
Raw OpenStreetMap data includes intermediate nodes placed along street curves for cartographic rendering. These points do not represent true intersections or dead-ends. OSMnx cleans the network topology by removing non-intersection nodes and consolidating intermediate sub-edges into single unified edges while preserving the complete spatial geometry and combined edge attributes.
Input: Directed multigraph $G = (V, E)$ constructed from raw OpenStreetMap ways
Input: Simplification mode $M \in \{\text{strict}, \text{non-strict}\}$
Output: Simplified multigraph $G' = (V', E')$
for each node $v \in V$ do
    is_endpoint $\leftarrow$ false
    if out_degree $(v) = 0$ or in_degree $(v) = 0$ then
        is_endpoint $\leftarrow$ true // Dead-end
    else if $v$ has a self-loop edge $(v, v)$ then
        is_endpoint $\leftarrow$ true
    else if $v$ represents a transition point between one-way and two-way streets then
        is_endpoint $\leftarrow$ true
    else
        incident_streets $\leftarrow$ set of physical undirected streets connected to $v$
        if $|incident_streets| \ge 3$ then
            is_endpoint $\leftarrow$ true // Multi-way intersection
        else if $|incident_streets| = 2$ then
            if $M = \text{strict}$ then
                is_endpoint $\leftarrow$ false // Considered a curve in a continuous road
            else if $M = \text{non-strict}$ then
                if incident streets have different OpenStreetMap IDs then
                    is_endpoint $\leftarrow$ true
                else
                    is_endpoint $\leftarrow$ false
                end if
            end if
        end if
    end if
    if is_endpoint is false then
        mark $v$ as non-intersection candidate for removal
    end if
end for
for each connected chain of candidate nodes between true endpoints do
    construct a unified edge connecting the true endpoints
    assign aggregated polyline geometry, summed length, and merged attributes to unified edge
    remove candidate nodes and intermediate sub-edges from $G$
end for
return $G'$
Knowl 4 — Definitions of Metric and Topological Street Network Measures in OSMnx
definition
OSMnx automates the calculation of structural and metric network indicators:
- Average street length ( $\bar{L}_{\text{street}}$ ): The mean edge length (in meters) in the undirected representation of the graph, serving as a proxy for block size.
- Intersection density ( $D_{\text{int}}$ ): Number of intersection nodes ( $n_{\text{int}}$ , defined as nodes with more than one physical street emanating from them) divided by the network land area $A$ in square kilometers: $D_{\text{int}} = n_{\text{int}} / A$ .
- Street density ( $D_{\text{street}}$ ): Sum of all physical street lengths in the undirected representation divided by land area $A$ (in $\text{km}/\text{km}^2$ ).
- Average circuity ( $\bar{C}$ ): Ratio of total physical edge length to the sum of straight-line great-circle distances between the endpoints of each edge:
$\bar{C} = \frac{\sum_{e \in E} \text{length}(e)}{\sum_{e=(u,v) \in E} d_{\text{great-circle}}(u, v)}$
- Average streets per node ( $\bar{k}_{\text{streets}}$ ): Mean number of physical undirected streets connected to each node (intersections and dead-ends), describing physical network connectivity independently of directed traffic rules.
- Average node connectivity ( $\bar{\kappa}$ ): The mean number of internally node-disjoint paths between all pairs of non-adjacent nodes in the network, representing the expected number of nodes that must fail to disconnect a randomly selected pair of non-adjacent nodes.
- Betweenness centrality ( $C_B(v)$ ): Fraction of shortest paths between all node pairs $(s, t)$ that traverse node $v$ :
$C_B(v) = \sum_{s \neq v \neq t} \frac{\sigma_{st}(v)}{\sigma_{st}}$

where $\sigma_{st}$ is the total count of shortest paths from $s$ to $t$ and $\sigma_{st}(v)$ is the count of those paths passing through $v$ .
- Closeness centrality ( $C_C(v)$ ): Reciprocal of the sum of length-weighted shortest-path distances from node $v$ to all other nodes in the network:
$C_C(v) = \frac{1}{\sum_{u \neq v} d_{\text{weighted}}(v, u)}$

Knowl 5 — Empirical Comparison of Street Network Topologies in Portland Neighborhoods

data/table

Three $0.5\text{ km}^2$ study areas in Portland, Oregon—Downtown (orthogonal 19th-century grid), Laurelhurst (early 20th-century streetcar suburb), and Northwest Heights (contemporary curvilinear suburban development)—were analyzed using OSMnx to demonstrate how urban form is captured by metric and topological network statistics.

Measure	Downtown	Laurelhurst	NW Heights
Area ( $\text{km}^2$ )	0.5	0.5	0.5
Intersection count	82	55	14
Intersection density ( $\text{km}^{-2}$ )	163.7	109.8	28.0
Total street length ( $\text{km}$ )	10.7	7.8	2.7
Street density ( $\text{km}/\text{km}^2$ )	21.3	15.6	5.4
Total edge length ( $\text{km}$ )	10.68	14.80	5.36
Edge density ( $\text{km}/\text{km}^2$ )	21.32	29.55	10.71
Average street segment length (m)	76.3	91.8	116.6
Average edge length (m)	76.3	97.4	116.6
Average streets per node	3.93	3.58	2.38
Average node degree	3.42	5.53	4.38
Average circuity	1.001	1.007	1.090
Diameter (m)	1278	1021	898
Radius (m)	742.9	537.1	561.8
Avg of the avg neighborhood degree	1.64	2.98	2.75
Avg of the avg weighted neighborhood degree	0.024	0.059	0.030
Avg clustering coefficient	$<0.001$	0.108	$<0.001$
Avg weighted clustering coefficient	$<0.001$	0.023	$<0.001$
Avg degree centrality	0.042	0.102	0.219
Avg closeness centrality	0.002	0.002	0.002
Avg betweenness centrality	0.070	0.077	0.137
Max PageRank value	0.030	0.029	0.106
Min PageRank value	0.002	0.004	0.017
Node connectivity	1	1	1
Edge connectivity	1	1	1
Avg node connectivity	1.326	2.107	1.443
Avg node connectivity (undirected)	2.868	2.496	1.443
Node density ( $\text{km}^{-2}$ )	163.7	109.8	41.9
Self-loop proportion	0	0	0
Street segment count	140	85	23
$n$ (nodes)	82	55	21
$m$ (edges)	140	152	46

The empirical results show that Downtown has the highest intersection density ( $163.7\text{ km}^{-2}$ ) and shortest average street segment length ( $76.3\text{ m}$ ), reflecting a fine-grained grid. Northwest Heights is coarse-grained and sparse, with low intersection density ( $28.0\text{ km}^{-2}$ ) and longer street segments ( $116.6\text{ m}$ ). In Downtown, total edge length matches total street length because every street is one-way, whereas in Laurelhurst bidirectional streets cause total edge length ( $14.80\text{ km}$ ) to nearly double total street length ( $7.8\text{ km}$ ).

Knowl 6 — Impact of Street Directionality on Topological Network Resilience
empirical result
Directionality constraints significantly reduce topological resilience and routing redundancy in dense urban street networks.

In the empirical case study of Portland, Downtown has a dense, fine-grained grid ( $163.7\text{ intersections/km}^2$ ) with predominantly 4-way intersections (averaging $3.93$ streets per node). However, because 100% of Downtown streets are one-way, its directed average node connectivity is only $1.326$ —lower than both Laurelhurst ( $2.107$ ) and Northwest Heights ( $1.443$ ).

When evaluated without directionality constraints (as an undirected graph), Downtown's average node connectivity rises to $2.868$ (more than doubling), exceeding Laurelhurst ( $2.496$ ) and Northwest Heights ( $1.443$ ). This indicates that the strict one-way configuration in Downtown drastically restricts the number of independent alternative paths between node pairs, suggesting that converting one-way streets to two-way streets would substantially improve network routing resilience for direction-constrained modes.
Knowl 7 — Betweenness Centrality Concentration and Chokepoint Vulnerability in Curvilinear Networks
empirical result
The spatial structure of street networks governs the concentration of network flow and systemic vulnerability to chokepoints:
- In an orthogonal grid (Downtown Portland), shortest paths are dispersed evenly across multiple parallel routes. Betweenness centrality is highest at the geographic center, and the single most critical node handles only $15\%$ of all shortest paths in the network (with an average node betweenness centrality of $0.070$ ).
- In a tree-like, curvilinear suburban network with cul-de-sacs (Northwest Heights), routing options are concentrated through a small number of connectors. The single most central node lies on $43\%$ of all shortest paths, and an average node carries $14\%$ of all shortest paths (average betweenness centrality of $0.137$ ).
Consequently, suburban and hierarchical street layouts exhibit severe vulnerability to single points of failure (e.g., traffic incidents, construction, or environmental hazards) relative to interconnected grid networks.
Knowl 8 — OSMnx Export and Visualization Capabilities
model/method
OSMnx provides native interfaces for network serialization, spatial data interchange, and cartographic visualization:
- File Serialization: Networks can be exported and loaded as GraphML files (preserving graph-theoretic and spatial attributes for use in NetworkX or Gephi) and as ESRI shapefiles. When exporting shapefiles, the graph is cast to an undirected representation while embedding one-way flags and origin/destination node IDs as tabular edge attributes to support external GIS routing.
- Vector Graphics: Networks can be saved directly as Scalable Vector Graphics (SVG) files for graphic editing.
- Visualization Functions: Includes built-in routines to plot:
  1. Shortest-path routes weighted by length, travel time, or elevation/grade impedance.
  2. Spatial distributions of specific node classes (such as dead-ends, 3-way, or 4-way intersections) to identify locations of structural disconnectivity.
  3. Standardized one-square-mile figure-ground diagrams to computationally compare urban fabric and morphology across cities.
Knowl 9 — OpenStreetMap Coverage and Attribute Limitations in OSMnx
limitation
OSMnx is constrained by the inherent properties and quality of OpenStreetMap crowdsourced data:
- Geographic Coverage Disparities: OpenStreetMap street network coverage is comprehensive across North America and Europe, but remains less complete or detailed in developing nations and rural areas.
- Pedestrian and Streetscape Attributes: While OpenStreetMap increasingly captures tags for sidewalks, street trees, lanes, and speed limits, qualitative attributes describing the experiential pedestrian environment and streetscape quality remain uneven across geographic regions.

Coverage note — The nationwide analysis of 27,000 U.S. street networks mentioned in the discussion and appendix was omitted as it represents a separate companion study.

References

1.Agryzkov, T., Oliver, J. L., Tortosa, L., & Vicent, J. F. (2012). An algorithm for ranking the nodes of an urban network based on the concept of PageRank vector. Applied Mathematics and Computation, 219(4), 2186–2193. doi:10.1016/j.amc.2012.08.064
2.Barron, C., Neis, P., & Zipf, A. (2014). A Comprehensive Framework for Intrinsic OpenStreetMap Quality Analysis. Transactions in GIS, 18(6), 877–895. doi:10.1111/tgis.12073
3.Barthélemy, M. (2004). Betweenness centrality in large complex networks. The European Physical Journal B, 38(2), 163–168. doi:10.1140/epjb/e2004-00111-4
4.Barthélemy, M. (2011). Spatial networks. Physics Reports, 499(1–3), 1–101. doi:10.1016/j.physrep.2010.11.002
5.Barthélemy, M., Bordin, P., Berestycki, H., & Gribaudi, M. (2013). Self-organization versus top-down planning in the evolution of a city. Scientific Reports, 3. doi:10.1038/srep02153
6.Barthélemy, M., & Flammini, A. (2008). Modeling Urban Street Patterns. Physical Review Letters, 100(13). doi:10.1103/PhysRevLett.100.138702
7.Beineke, L. W., Oellermann, O. R., & Pippert, R. E. (2002). The Average Connectivity of a Graph. Discrete Mathematics, 252(1), 31–45. doi:10.1016/S0012-365X(01)00180-7
8.Boeing, G. (2017). A Multi-Scale Analysis of 27,000 Urban Street Networks. Manuscript submitted for publication. doi:10.2139/ssrn.2943038
9.Brin, S., & Page, L. (1998). The anatomy of a large-scale hypertextual web search engine. Computer Networks and ISDN Systems, 30(1–7), 107–117. doi:10.1016/S0169-7552(98)00110-X
10.Buhl, J., Gautrais, J., Reeves, N., Solé, R. V., Valverde, S., Kuntz, P., & Theraulaz, G. (2006). Topological patterns in street networks of self-organized urban settlements. The European Physical Journal B, 49(4), 513–522. doi:10.1140/epjb/e2006-00085-1
11.Cardillo, A., Scellato, S., Latora, V., & Porta, S. (2006). Structural properties of planar graphs of urban street patterns. Physical Review E, 73(6). doi:10.1103/PhysRevE.73.066107
12.Cervero, R., & Kockelman, K. (1997). Travel Demand and the 3 Ds: Density, Diversity, and Design. Transportation Research Part D, 2(3), 199–219. doi:10.1016/S1361-9209(97)00009-6
13.Chin, W.-C.-B., & Wen, T.-H. (2015). Geographically Modified PageRank Algorithms. PLoS ONE, 10(10), e0139509. doi:10.1371/journal.pone.0139509
14.Corcoran, P., Mooney, P., & Bertolotto, M. (2013). Analysing the growth of OpenStreetMap networks. Spatial Statistics, 3, 21–32. doi:10.1016/j.spasta.2013.01.002
15.Crucitti, P., Latora, V., & Porta, S. (2006). Centrality measures in spatial networks of urban streets. Physical Review E, 73(3), 036125. doi:10.1103/PhysRevE.73.036125
16.Ewing, R., & Cervero, R. (2010). Travel and the Built Environment: A Meta-Analysis. Journal of the American Planning Association, 76(3), 265–294. doi:10.1080/01944361003766766
17.Foti, F. (2014). Behavioral Framework for Measuring Walkability and its Impact on Home Values and Residential Location Choices (Dissertation). University of California, Berkeley, CA.
18.Frizzelle, B., Evenson, K., Rodriguez, D., & Laraia, B. (2009). The importance of accurate road data for spatial applications in public health: customizing a road network. International Journal of Health Geographics, 8(24). doi:10.1186/1476-072X-8-24
19.Giacomin, D. J., & Levinson, D. M. (2015). Road network circuity in metropolitan areas. Environment and Planning B, 42(6), 1040–1053. doi:10.1068/b130131p
20.Girres, J.-F., & Touya, G. (2010). Quality Assessment of the French OpenStreetMap Dataset. Transactions in GIS, 14(4), 435–459. doi:10.1111/j.1467-9671.2010.01203.x
21.Grammenos, F., & Pollard, D. (2009, October 19). Beloved and Abandoned: A Platting Named Portland. Planetizen.
22.Guzowski, K. J. (1990). Portland’s Olmsted Vision (1897-1915) (Thesis). University of Oregon, Eugene, OR.
23.Hagberg, A., & Conway, D. (2010). Hacking social networks using the Python programming language. Presented at the Sunbelt 2010, Riva del Garda, Italy.
24.Hajrasouliha, A., & Yin, L. (2015). The impact of street network connectivity on pedestrian volume. Urban Studies, 52(13), 2483–2497. doi:10.1177/0042098014544763
25.Haklay, M. (2010). How Good is Volunteered Geographical Information? A Comparative Study of OpenStreetMap and Ordnance Survey Datasets. Environment and Planning B, 37(4), 682–703. doi:10.1068/b35097
26.Hillier, B., Leaman, A., Stansall, P., & Bedford, M. (1976). Space Syntax. Environment and Planning B, 3(2), 147–185. doi:10.1068/b030147
27.Jacobs, A. (1995). Great Streets. Cambridge, MA: MIT Press.
28.Jiang, B. (2007). A Topological Pattern of Urban Street Networks: Universality and Peculiarity. Physica A, 384(2), 647–655. doi:10.1016/j.physa.2007.05.064
29.Jiang, B., & Claramunt, C. (2002). Integration of space syntax into GIS: new perspectives for urban morphology. Transactions in GIS, 6(3), 295–309. doi:10.1111/1467-9671.00112
30.Jiang, B., & Claramunt, C. (2004). Topological Analysis of Urban Street Networks. Environment and Planning B, 31(1), 151–162. doi:10.1068/b306
31.Jokar Arsanjani, J., Zipf, A., Mooney, P., & Helbich, M. (Eds.). (2015). OpenStreetMap in GIScience. Cham, Switzerland: Springer.
32.Karduni, A., Kermanshah, A., & Derrible, S. (2016). A protocol to convert spatial polyline data to network formats and applications to world urban road networks. Scientific Data, 3, 160046. doi:10.1038/sdata.2016.46
33.Karimi, K. (2012). A configurational approach to analytical urban design: “Space syntax” methodology. Urban Design International, 17(4), 297–318. doi:10.1057/udi.2012.19
34.Maron, M. (2015, November 19). How complete is OpenStreetMap? Mapbox.
35.Marshall, W., & Garrick, N. (2010). Street network types and road safety: A study of 24 California cities. Urban Design International, 15(3), 133–147. doi:10.1057/udi.2009.31
36.Masucci, A. P., Smith, D., Crooks, A., & Batty, M. (2009). Random planar graphs and the London street network. The European Physical Journal B, 71(2), 259–271. doi:10.1140/epjb/e2009-00290-4
37.Neis, P., Zielstra, D., & Zipf, A. (2011). The Street Network Evolution of Crowdsourced Maps: OpenStreetMap in Germany 2007–2011. Future Internet, 4(1), 1–21. doi:10.3390/fi4010001
38.Newman, M. E. J. (2010). Networks: An Introduction. Oxford, England: Oxford University Press.
39.O’Sullivan, D. (2014). Spatial Network Analysis. In M. M. Fischer & P. Nijkamp (Eds.), Handbook of Regional Science (pp. 1253–1273). Berlin: Springer. doi:10.1007/978-3-642-23430-9_67
40.Parthasarathi, P., Levinson, D., & Hochmair, H. (2013). Network Structure and Travel Time Perception. PLoS ONE, 8(10), e77718. doi:10.1371/journal.pone.0077718
41.Porta, S., Crucitti, P., & Latora, V. (2006). The network analysis of urban streets: a primal approach. Environment and Planning B, 33(5), 705–725. doi:10.1068/b32045
42.Ratti, C. (2004). Space syntax: some inconsistencies. Environment and Planning B, 31(4), 487–499. doi:10.1068/b3019
43.Sevtsuk, A., & Mekonnen, M. (2012). Urban network analysis. A new toolbox for ArcGIS. Revue Internationale de Géomatique, 22(2), 287–305. doi:10.3166/rig.22.287-305
44.Snyder, E. E. (1979). Portland Names and Neighborhoods: Their Historic Origins. Portland, OR: Binford & Mort.
45.Southworth, M., & Ben-Joseph, E. (1997). Streets and the Shaping of Towns and Cities. New York, NY: McGraw-Hill.
46.Strano, E., Viana, M., da Fontoura Costa, L., Cardillo, A., Porta, S., & Latora, V. (2013). Urban Street Networks, a Comparative Analysis of Ten European Cities. Environment and Planning B, 40(6), 1071–1086. doi:10.1068/b38216
47.Trudeau, R. J. (1994). Introduction to Graph Theory (2nd ed.). New York, NY: Dover Publications.
48.Works, M. (2016). Laurelhurst. In The Oregon Encyclopedia. Portland, OR: The Oregon Historical Society.
49.Zielstra, D., Hochmair, H. H., & Neis, P. (2013). Assessing the Effect of Data Imports on the Completeness of OpenStreetMap – A United States Case Study. Transactions in GIS, 17(3), 315–334. doi:10.1111/tgis.12037

Citation

MLA

Boeing, G. “OSMnx: New Methods for Acquiring, Constructing, Analyzing, and Visualizing Complex Street Networks”. Computers, Environment and Urban Systems, vol. 65, 2017, pp. 126–39, https://doi.org/10.1016/j.compenvurbsys.2017.05.004.

APA

Boeing, G. (2017). OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks. Computers, Environment and Urban Systems, 65, 126–139. https://doi.org/10.1016/j.compenvurbsys.2017.05.004

Chicago

Boeing, G. 2017. “OSMnx: New Methods for Acquiring, Constructing, Analyzing, and Visualizing Complex Street Networks”. Computers, Environment and Urban Systems 65: 126–39. https://doi.org/10.1016/j.compenvurbsys.2017.05.004.

Harvard

Boeing, G. (2017) “OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks”, Computers, Environment and Urban Systems, 65, pp. 126–139. Available at: https://doi.org/10.1016/j.compenvurbsys.2017.05.004.

Vancouver

1. Boeing G (2017) OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks. Computers, Environment and Urban Systems 65:126–139

BibTeX

@article{Boeing_2017, title={OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks}, volume={65}, ISSN={0198-9715}, url={http://dx.doi.org/10.1016/j.compenvurbsys.2017.05.004}, DOI={10.1016/j.compenvurbsys.2017.05.004}, journal={Computers, Environment and Urban Systems}, publisher={Elsevier BV}, author={Boeing, Geoff}, year={2017}, month=Sept, pages={126–139} }

Metadata:Crossref

Source Code

This paper has an official code repository available. Click below to access the source code.

View Repository

Access the Paper

This paper is available from its original source. Click below to access the PDF.

Open PDF

OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks

Abstract

Topics

Keywords

Table of Contents

Knowls

Knowl 1 — OSMnx Graph Representation of Urban Street Networks

Knowl 2 — Spatial Data Querying, Buffering, and Truncation Pipeline

Knowl 3 — Algorithmic Network Topological Simplification and Node Filtering

Knowl 4 — Definitions of Metric and Topological Street Network Measures in OSMnx

Knowl 5 — Empirical Comparison of Street Network Topologies in Portland Neighborhoods

Knowl 6 — Impact of Street Directionality on Topological Network Resilience

Knowl 7 — Betweenness Centrality Concentration and Chokepoint Vulnerability in Curvilinear Networks

Knowl 8 — OSMnx Export and Visualization Capabilities

Knowl 9 — OpenStreetMap Coverage and Attribute Limitations in OSMnx

References

Citation

BibTeX

Source Code

Access the Paper