MSRFormer: road network representation learning using multi-scale feature fusion of heterogeneous spatial interactions

Road network connectivity is an important indicator for measuring the operational efficiency and reliability of urban road networks, and it plays an important role in supporting traffic planning and management decisions. The implementation of traffic management measures, such as traffic bans and temporary traffic flow changes, will restrict access to some sections and lanes, reduce the passable paths in the road network, and thus affect the overall connectivity performance of the road network. Existing road network research results mostly evaluate the topological connectivity of the network at the physical level, and it is difficult to accurately portray the actual road network connectivity under traffic management conditions. To quantitatively evaluate the road network connectivity performance after the implementation of traffic management tools, this paper proposes a road network connectivity evaluation method based on strongly connected effective paths. Firstly, the node steering coefficients are used to describe the no‐traffic constraints of turning lanes, and the connectivity evaluation indexes are constructed based on the number of strongly connected effective paths and the shortest paths of strongly connected paths. Secondly, combining the Floyd‐Warshall algorithm and the depth‐first search algorithm, we constructed a strong connectivity effective path search algorithm to adapt to the refined traffic management situation, and identified the key road sections that have the greatest impact on the connectivity of the road network by considering the maximum acceptable level of the path and the road access constraints. Finally, Sioux‐Falls network and nine urban road networks with different layout patterns are selected for the case study and compared with traditional road network connectivity indicators. The case studies show that: (1) the connectivity of the square grid road network structure is superior, while the connectivity of the free‐form road network is the lowest; (2) road access management measures reduce the overall road network connectivity, and the banning of traffic in critical sections has the most significant effect on connectivity. Accurately assessing the changes in road network connectivity performance under different traffic management measures provides a scientific basis for the development of road control strategies, which can effectively improve urban traffic fluency and residents’ travel efficiency.

The urban road network shares self-similarity and is suitable to adopt the fractal theory for analysis. The fractal dimension of the urban road network reflects the completeness of the road system. The employment of fractal theory to analyze the urban road network can reflect the feasibility of the urban road network layout and is of vital significance to the evaluation of the urban road network planning. GIS has strong space analysis function and can directly show the analysis process. To use the fractal theory to analyze the urban road network on the GIS platform can provide a new idea for the evaluation of the urban road network. Based on the GIS platform, this paper employs the fractal theory to the analysis of Shenzhen's municipal road network. Hausdorff method is adopted to calculate the index of the road network coverage, analyze the calculation results and put forward suggestions to optimize the road network.

Identifying spatial interaction patterns of vehicle movements on urban road networks by topic modelling

The reasonable connection between the external freeway network and the urban road network can alleviate the traffic pressure in the city. The fusion evaluation between the external freeway network and the urban road network are studied. Firstly, considering from three aspects of the facilities and the service performance and the connection point operation quality, the evaluation indicator system of two networks is established. Then, based on the cloud model and fuzzy comprehensive evaluation method, the fusion evaluation method based on cloud model - fuzzy comprehensive evaluation method of two networks is established, and the fusion evaluation process is determined. Finally, the effectiveness and scientific of the fusion evaluation method are verified by a case analysis of the skeleton road network and the external freeway of Changsha city.

Link travel speeds in road networks are essential data for a variety of research problems in logistics, transportation, and traffic management. Real-world link travel speeds are stochastic, and highly dependent on speeds in previous time periods and neighboring road links. To understand how link travel speeds vary over space and time, we uncover their distributions, their space- and/or time-dependent correlations, as well as partial correlations, based on link travel speed datasets from an urban road network and a freeway network. We find that more than 90% (57%) of travel speeds are normally distributed in the urban road (freeway) network, and that correlations generally decrease with increased distance in time and space. We also investigate if and how different types of road links affect marginal distributions and correlations. The results show that different road link types produce quite similar marginal distributions and correlations. Finally, we study marginal distributions and correlations in a freeway network. Except that the marginal distribution and time correlation are different from the urban road network, others are similar.

The city development is closely related to the performance of the transportation network system. Bridges and roads are important parts of the transportation system, and are also inseparable components of the transportation network. However, the effect of the correlation between bridges and roads on the network system has not been studies thoroughly in the literature. Therefore, it is necessary to analyse the vulnerability of the road network when both bridges and roads are involved. In this paper, the urban road network is modeled into the form of network connection and node, based on the analysis of the related research results of road network vulnerability in the literature. Taking the urban roads at all levels as the connection and the transportation hubs (including bridges) as the nodes, the paper puts forward the corresponding measurement indexes and calculation methods, and establishes the importance and correlation analysis model of roads and bridges in the urban road network. At last, the model is applied to the road network which is 5 × 3 km2 besides Yangpu Bridge of Shanghai for verification, the importance and correlation of specific roads and bridges in the analyzed urban road network are calculated, which provides a certain basis for dealing with various emergencies leading to the decline of urban road network vulnerability. In this paper, the importance analysis of urban road network is extended to the bridge correlation analysis, so that the proposed model of the vulnerability assessment of the urban road network system is more suitable for the increasingly demand of road and bridge construction in China, and provides a certain basis for dealing with the decline of road network vulnerability caused by various emergencies.

The vulnerability of an urban road network is affected by many factors, such as internal road network layout, network structure strength, and external destructive events, which have great uncertainty and complexity. Thus, there is still no unified and definite vulnerability analysis scheme available to cities. This paper proposes an integrative vulnerability identification method for urban road networks, which mainly relates to the vulnerability connotation and characteristics analysis of urban road networks during emergency, and vulnerability comprehensive evaluation indices design based on urban road network connectivity, traffic efficiency and performance, and an empirical study on a vulnerability identification method of an urban road network. In the empirical case, a real road network and traffic operation data were used from Science and Technology Park of Shenzhen City, China. In the context of one certain emergency scenario, the stated preference survey method and maximum likelihood method are used to solve the road users’ random travel choice behavior parameters; subsequently, based on the traffic equilibrium distribution prediction, the traffic vulnerability identification methods of the road network in this region were verified before and after the emergency. The method presented here not only considers the impact of network topology changes on road network traffic function during emergency but also considers the impact of dynamic changes in road network traffic demand on vulnerability; therefore, it is closer to the actual distribution of urban road network traffic vulnerability.

A Traffic Congestion Assessment Method for Urban Road Networks Based on Speed Performance Index

OpenStreetMap offers a valuable source of worldwide geospatial data useful to urban researchers. This study uses the OSMnx software to automatically download and analyze 27,000 US street networks from OpenStreetMap at metropolitan, municipal, and neighborhood scales—namely, every US city and town, census urbanized area, and Zillow-defined neighborhood. It presents empirical findings on US urban form and street network characteristics, emphasizing measures relevant to graph theory, transportation, urban design, and morphology such as structure, connectedness, density, centrality, and resilience. In the past, street network data acquisition and processing have been challenging and ad hoc. This study illustrates the use of OSMnx and OpenStreetMap to consistently conduct street network analysis with extremely large sample sizes, with clearly defined network definitions and extents for reproducibility, and using nonplanar, directed graphs. These street networks and measures data have been shared in a public repository for other researchers to use.

OpenStreetMap offers a valuable source of worldwide geospatial data useful to urban researchers. This study uses the OSMnx software to automatically download and analyze 27,000 US street networks from OpenStreetMap at metropolitan, municipal, and neighborhood scales - namely, every US city and town, census urbanized area, and Zillow-defined neighborhood. It presents empirical findings on US urban form and street network characteristics, emphasizing measures relevant to graph theory, transportation, urban design, and morphology such as structure, connectedness, density, centrality, and resilience. In the past, street network data acquisition and processing have been challenging and ad hoc. This study illustrates the use of OSMnx and OpenStreetMap to consistently conduct street network analysis with extremely large sample sizes, with clearly defined network definitions and extents for reproducibility, and using nonplanar, directed graphs. These street networks and measures data have been shared in a public repository for other researchers to use.

Identifying correspondences of road segments in different road networks, namely road-network matching, is an essential task for road network-centric data processing such as data integration of road networks and data quality assessment of crowd-sourced road networks. Traditional road-network matching usually relies on feature engineering and parameter selection of the geometry and topology of road networks for similarity measurement, resulting in poor performance when dealing with dense and irregular road network structures. Recent development of graph neural networks (GNNs) has demonstrated unsupervised modeling power on road network data, which learn the embedded vector representation of road networks through spatial feature induction and topology-based neighbor aggregation. However, weighting spatial information on the node feature alone fails to give full play to the expressive power of GNNs. To this end, this paper proposes a Spatial Pattern-aware Graph EMbedding learning method for road-network matching, named SP-GEM, which explores the idea of spatially-explicit modeling by identifying spatial patterns in neighbor aggregation. Firstly, a road graph is constructed from the road network data, and geometric, topological features are extracted as node features of the road graph. Then, four spatial patterns, including grid, high branching degree, irregular grid, and circuitous, are modelled in a sector-based road neighborhood for road embedding. Finally, the similarity of road embedding is used to find data correspondences between road networks. We conduct an algorithmic accuracy test to verify the effectiveness of SP-GEM on OSM and Tele Atlas data. The algorithmic accuracy experiments show that SP-GEM improves the matching accuracy and recall by at least 6.7% and 10.2% among the baselines, with high matching success rate (>70%), and improves the matching accuracy and recall by at least 17.7% and 17.0%, compared to the baseline GNNs, without spatially-explicit modeling. Further embedding analysis also verifies the effectiveness of the induction of spatial patterns. This study not only provides an effective and practical algorithm for road-network matching, but also serves as a test bed in exploring the role of spatially-explicit modeling in GNN-based road network modeling. The experimental performances of SP-GEM illuminate the path to develop GeoEmbedding services for geospatial applications.

Interaction between Road Network Connectivity and Spatial Pattern

Exploring the relationship between the spatial distribution of roads and universal pattern of travel-route efficiency in urban road networks

Analysis of street networks is a challenging task, needed in urban planning applications such as urban design or transportation network analysis. Typically, different network features of interest are used for within- and between comparisons across street networks. We introduce StreetExplorer, a visual-interactive system for analysis and comparison of global and local patterns in urban street networks. The system uses appropriate similarity functions to search for patterns, taking into account topological and geometric features of a street network. We enhance the visual comparison of street network patterns by a suitable color-mapping and boosting scheme to visualize the similarity between street network portions and the distribution of network features. Together with experts from the urban morphology domain, we apply our approach to analyze and compare two urban street networks, identifying patterns of historic development and modern planning approaches, demonstrating the usefulness of StreetExplorer.

Quantifying the topological similarities of different parts of urban road networks enables us to understand urban growth patterns. Although conventional statistics provide useful information about the characteristics of either a single node’s direct neighbours or the entire network, such metrics fail to measure the similarities of subnetworks or capture local, indirect neighbourhood relationships. Here we propose a graph-based machine learning method to quantify the spatial homogeneity of subnetworks. We apply the method to 11,790 urban road networks across 30 cities worldwide to measure the spatial homogeneity of road networks within each city and across different cities. We find that intracity spatial homogeneity is highly associated with socioeconomic status indicators such as gross domestic product and population growth. Moreover, intercity spatial homogeneity values obtained by transferring the model across different cities reveal the intercity similarity of urban network structures originating in Europe, passed on to cities in the United States and Asia. The socioeconomic development and intercity similarity revealed using our method can be leveraged to understand and transfer insights between cities. It also enables us to address urban policy challenges including network planning in rapidly urbanizing areas and regional inequality. The spatial homogeneity of urban road networks can be quantified in a fine-grained manner with graph neural networks. This method is studied across 11,790 inner-city road networks around the world and can be used to study socioeconomic development and help with urban planning.