Advances in geospatial approaches to transport networks and sustainable mobility

Abstract

Introduction

Transport plays a pivotal role within urban environments, shaping the capacity for the efficient and resilient movement of people and goods. In doing so, it influences the economic productivity and environmental sustainability of cities. The intricate relationship between the spatial distribution of various land uses and the transport modes employed is a defining factor in a city's economic functionality, as already recognised long ago (Von Thünen, 1826). Nevertheless, this intricate interaction has posed formidable challenges to resolve, given that transport also gives rise to issues such as congestion, pollution, inequality, and dependence on fossil fuels. Addressing these challenges and seeking more sustainable solutions necessitates the collaboration of diverse disciplines. Geographers, engineers, physicists, planners, social scientists, data scientists, and others must come together to optimise transport and mobility systems while carefully considering their societal and environmental implications. Overall, the aim of this special issue was to bring together this diverse range of disciplines and present some of the recent advances in geospatial approaches for studying transport networks and sustainable mobility.

Let us start by identifying the fundamental concepts that are instrumental for the understanding of the dynamics and the processes intertwining the transport systems and other urban dimensions. First and foremost is accessibility, which refers to the ease with which goods, services, and activities can be reached. There are many nuances in how accessibility is defined and operationalised in the scientific literature (see, e.g. Handy and Niemeier, 1997; Levinson and Wu, 2020), but in essence, they all conceive of it as an essential component of many urban processes in which people, transport, and land uses are in continuous interaction with each other. Another fundamental concept closely related to accessibility is mobility, which refers to the actual movement of individuals between places and is often a function of both the means and ease of travel (Giuliano and Hanson, 2017). Connectivity, meanwhile, is the degree to which different locations or nodes are linked, shedding light on the structural intricacies and interdependencies of transport. Anchoring these concepts together is the notion of transport networks, which comprises the physical and operational structures, be it roads, railways, airways, or waterways, that facilitate the movement of people and goods. These fundamental concepts interweave to form the intricate tapestry of transport systems and their pivotal role in shaping modern-day societies and economies.

In the era of Big Data and increased modelling capabilities, the study of transport networks and mobility systems has expanded rapidly among quantitative researchers. The emergence of various novel data sources including mobile phone data, GPS-devices, social media data, and different sensors has enabled novel approaches for studying mobility, spatial accessibility and transport related questions with a level of detail and scale that was merely a distant dream a few decades ago. Interest in data-driven approaches for studying transport systems has exploded in the past decade or so. This rapid change in the way the research is conducted has also raised some concerns. For example, Schwanen (2017) points out that although the latest advancements in the availability of data and interest in data-driven approaches bring many new possibilities to study mobility systems, we should be careful not to ‘privilege the generality over particularity’ in the transport research, meaning that the local contexts, time and particular situations should be considered and reflected when analysing transport related phenomena. From our perspective, we see this cross-disciplinary interest in transport as a good opportunity for learning about transport and mobility from various perspectives. With this special issue, we particularly wanted to invite scientific contributions with a broad (quantitative) scope, covering approaches stemming from more ‘traditional’ transport geography, as well as approaches from urban/complexity/network science. The different approaches can then provide complementary insights and learnings that will hopefully help to advance science and the understanding of our mobility systems.

This editorial is organised as follows. It initially identifies the different methodological approaches to transport research, then provides an overview of the different contributing papers, and finally, it highlights the opportunities to further develop this field.

Methodological diversity in transport research

Societies are facing many complex challenges such as climate change and growing urban and social inequalities in which transport plays a key role. Tackling these grand challenges requires crossing disciplinary boundaries and developing novel measures and methods that bring new insights into the structure and dynamics of transport networks on multiple levels (from individual to system level). In this special issue, we cover articles with a large variety of (mostly open) data and methods that stem from geographic information science and network science, which are applied to understanding transport and mobility from multiple perspectives. Our collection contains multiple studies that use and have developed techniques to study e.g. GPS-trajectories, origin-destination data, and street network data to understand sustainable mobility and transport-related questions. The issue also features new advances in using machine learning to explore regional development based on longitudinal time-series data, as well as simulation techniques to understand navigation and route-choice related questions. The following section provides a more comprehensive overview of the different methodological approaches used by the authors contributing to this special issue.

What did we learn from this special issue?

This special issue consists of 14 articles that focus on different aspects, such as modelling approaches to cycling (5 articles, the most popular topic), motorised vehicles and driving (4), transit (2), railways (2), pedestrians (2), and generic street networks (1). Almost half (43%) of the papers studied transport from a multimodal perspective that integrates more than one transport mode in their analysis. The majority of the papers focused on studying transport in urban areas (79% of the papers) in Europe (64%), while individual studies were conducted in Africa, South America, and North America. One of the studies had a comparative international study design covering cities from four continents (Europe, America, Asia, and Oceania), and one study was a review paper without a specific geographical scope. Half of the studies (7 papers) used mostly open data sources, such as OpenStreetMap or open data from national authorities. In the following, we specify the main outcomes of this collection.

Learning from the general – scaling of transport infrastructure

Interest in scaling laws with a more general view of transport can provide valuable lessons when multiple cities worldwide are studied simultaneously in a comparable manner. For example, Reggiani et al. (2023) in this special issue discovered that larger cities tend to systematically invest less in cycling infrastructure based on a comparison of 47 cities across four continents. They observed that cities that have double the population appear to have 24% more kilometres of bicycle dedicated infrastructure, but around 80% more car road infrastructure. These findings provide an important lesson and emphasises the need to do more in larger cities to support low-carbon transport and sustainable mobility, for example, by building more cycling infrastructure and improving existing residential roads to make them safer to cycle.

Approaches to study transport on multimodal networks

Understanding various mobility related questions requires appropriate methods and tools, which most often rely on multimodal analyses that consider the interplay between different transport modes. Alessandretti et al. (2023) provide a comprehensive overview of the various open-source tools and approaches for studying urban multimodal mobility and transport infrastructure as multilayer networks. They show that important advances in multimodal mobility and transport have been interdisciplinary, and studying these types of questions has clearly benefited from the large variety of scientific fields and practices. However, they also highlight that existing research on multimodal travel based on passively collected data sources is far from being comprehensive. Most empirical studies on multimodal behaviour have focused on the use of public transport (combining transit and walking), but there seems to be a gap in research that aims to understand the interplay between forms of public and private transport such as walking, driving and cycling using multilayer networks.

Lunke et al. (2023) have studied multimodal accessibility and they compare the competitiveness of public transport against driving in 13 urban regions in Norway. In their accessibility comparisons, they also take economic and geographical perspectives into account. This approach provides a more realistic and comprehensive view of the competitiveness of public transport compared to access by private car. Their methodological approach demonstrates the need to incorporate quality factors, such as crowdedness, congestion, transfers and waiting times, into accessibility analyses in addition to travel time, which is by far the most widely used metric in accessibility studies.

Developing cycling infrastructure with data-driven approaches

New approaches to help planning for better cycling infrastructure are actively being constructed. For instance, Folco et al. (2023) developed a data-driven method combining empirical micro-mobility data and a network growth model (Szell et al., 2022) to simulate how improvements to existing cycling infrastructure align with the demand, and how possible investments to infrastructure influence safety as the network grows. In addition, Beecham et al. (2023) developed a connected bikeability index that considers safety, comfort, attractiveness, and coherence, to understand whether different parts of London, UK, are better connected by bike than others. Based on their bikeability index, they found a connection between bikeability levels and the hierarchy of city areas: the more strategic importance there is between specific areas/villages in London, the higher the bikeability index. Based on insights from the city bike system data, they also found evidence that higher bikeability tends to increase the cycling activity, especially if there are many jobs in the neighbourhood.

In addition to safety and connectivity related questions, it is important to understand how citizens use the cycling networks based on their preferences, in order to better design and plan for cycling infrastructure. Aras et al. (2023) investigated the role of intersection crossings as physical barriers to cycling based on bikeshare data from Chicago, USA. They found that bikeshare riders prefer avoiding large intersections in circumstances when it is possible to do so. Also De Jong et al. (2023) identified that cyclists prefer routes with fewer crossings. Based on a large sample of GPS-trajectories in Oslo, Norway, they studied which factors influence cycling route choices. They found that cyclists tend not to use the shortest paths when travelling between locations, but they prefer and value good cycling infrastructure, flatter routes as well as routes that are close to water areas. Somewhat surprisingly, they also found that routes passing through green spaces are cycled less often than expected.

Understanding the uncertainties in travel

Understanding uncertainties related to quantitative research is an evergreen topic (Franklin, 2022), and highly relevant to navigation and transport, which are inherently stochastic by nature. Three papers in this special issue advance the science in this domain by incorporating uncertainties and variability in navigation and public transport accessibility. Thompson Sargoni and Manley (2023) developed a novel hierarchical pedestrian route choice framework based on construal level theory (Trope and Liberman, 2012) that integrates dynamic, perceptual decisions at the street level with abstract, and network-based decisions at the neighbourhood level. They show with a case-study focusing on pedestrian navigation in London, UK, that movement is affected by decision making at both the neighbourhood and street level. Thus, when modelling pedestrian movement, considering both street-level and neighbourhood level decisions can support more detailed assessments of pedestrian walking experienced in cities. Corcoran and Lewis (2023) conducted a larger-scale analysis in the UK, and developed an approach to estimate how the navigability of streets varies in 48 cities. They assessed the entropy of street networks on a local and global level, which can be used to estimate how difficult it is to navigate on a given street network. This approach can provide useful information for urban planners to identify cities that are more difficult to navigate and therefore might need additional signage to improve the navigability in the city.

Stewart and Byrd (2023) studied the uncertainties and variations in travel times and public transport accessibility indicators. They compared how public transport waiting times vary between a conventional half-headway method and a more advanced Monte Carlo method developed for the R5 routing engine (Conway et al., 2017). By conducting an empirical study in Santiago, Chile, they found that the conventional method tends to underestimate the benefits of public transport in certain locations, particularly those served by multiple transit lines (Stewart and Byrd, 2023). This can have implications for planning and practice, and they recommend using the Monte Carlo method when analysing complex networks or comparing transport scenarios.

Characterising the transport networks and their dynamics

To better comprehend how our dynamic mobility systems work, it is important to understand the features and characteristics of the built environment that make the movement along the streets possible. Palominos and Smith (2022) introduce a novel streetspace allocation analysis method that uses street cross-sections to measure footway and carriageway widths of the street network. This approach was used to evaluate citywide street design in London, UK, and they show that the technique can provide greater insights into understanding the street network structure, and connectivity patterns. By doing so, they were able to gain insights into the functioning of the pedestrian street network. Their method can offer valuable insights to practical planning and analytical capacity for developing new transit-oriented schemes or designing place-based streets supporting sustainable transport and urban development.

The study by Bruwer and Andersen (2023) focuses on understanding the dynamics of connectivity and traffic flow in Stellenbosch, South Africa. They used floating car data to study COVID-19 related traffic changes and systematically evaluated various indicators that measure traffic congestion on street networks. They found the Speed Reduction Index to be the most suitable metric to evaluate spatial and temporal congestion patterns, and interestingly, that the congestion level along arterial roads is a function of flow in both directions: high flow in one direction correlates with increasing congestion in the other direction.

Transport as a driver of regional development

This issue introduces two papers that study the connection between transport infrastructure and regional development in two European countries. Lai et al. (2022) studied how the Swiss transport infrastructure has evolved in the 20th Century (1910–2000), and how changes in the infrastructure have influenced regional development in Switzerland, covering 2833 municipalities. They used a deep learning approach (Graph Convolutional Networks and SHAP) to predict regional development considering population and rail/road network density as development indicators. Their approach provides valuable insights into understanding the history of regional development and illustrates the changing population dynamics in Switzerland. They show that predicting population growth (or decline) requires a good understanding of the city’s connectedness, and demonstrate that higher connectivity in specific areas can lead to population decline or increase in the surrounding municipalities, depending on the transport system characteristics. Their results illustrate that a well-developed railway network decreases the tendency to move away from a given municipality, while better road accessibility tends to increase the likelihood of moving out of large cities, as it is easier to commute longer distances (Lai et al., 2022).

Prignano et al. (2023) proposed a methodology inspired by archaeological research and network science to shed light on the processes and forces that have moulded transport infrastructures into their current configuration. By using a mechanistic network model, the authors generate several synthetic transport networks with different parameters and compare the outputs against the empirical real-world passenger transport network in Spain. The approach can provide useful insights to understand and model how transport networks evolve under certain circumstances.

Blindspots: What is overlooked?

As many scholars before us have identified (e.g. Hanson, 2006; Lowe, 2021; Schwanen, 2017; Willberg et al., 2023), the scientific endeavour focusing on transport tends to follow specific trends. Hanson (2006) already challenged transport researchers two decades ago to ‘imagine questions, methodologies, and epistemologies’ beyond the most obvious ones. Lowe (2021) identified that, for example, targeted funding, researchers’ positionalities and disciplinary traditions largely dictate what is studied (and where), and what is left out of the radar. Based on the themes of the articles published in this special issue, we can also say something about the areas that tend to receive little attention in quantitative transport research.

First, the geographical focus of the studies in our special issue is heavily biased towards Europe. Naturally, our special issue only covers a small sample of all research, but a similar geographical bias in transport research towards the more affluent countries of the global economy (Europe, North America, Australia and China) is well known based on systematic reviews (Derudder et al., 2019). Thus, in line with Schwanen (2018), we call for more quantitative transport research that would expand beyond its traditional heartlands to Africa, Latin America, and Asia. Preferably, this research would also go beyond reinforcing ‘Western’ concepts, tools, data, and current understanding about transport in these understudied areas (Schwanen, 2017).

Second, the special issue aimed to attract papers that would study the complexity and resilience of multimodal transport networks or use alternative sustainability-focused indicators and measures to study transport related questions from a socio-ecological perspective (using, e.g. environmental costs, exposures, or monetary costs). Based on the coverage of the collection (14 papers), as well as recent reviews (Dillman et al., 2021; Willberg et al., 2023), it is evident that the socio-ecological aspects of transport/mobility/accessibility still receive relatively little attention in quantitative transport research. With the climate emergency at hand, the transport and mobility systems need to be transformed to meet climate goals and reduce negative environmental and social effects. Thus, we see a lot of potential and need for future scientific advancements in this area from the transport studies and network science perspective.

Lastly, the focus of the research in this collection tends to be on relatively short-distance travel in urban areas (11 out of 14 studies), overlooking studies on rural areas (Poorthuis and Zook, 2023) as well as regional or long-distance travel (Heinen and Mattioli, 2019). The focus on short-distance travel in urban areas is not surprising, since there are many drivers that make it easier for researchers to focus on transport/mobility that happens within city regions. For instance, (i) getting access to relevant and high-quality data is easier from more limited geographical areas (compared to, e.g. intercity travel), (ii) the uncertainties in the data and complexity in the models are easier to handle, and (iii) the computational resources needed for conducting the analyses are lower. Naturally, we scientists should aim to push the boundaries of what is possible to do. Hence, we call for more quantitative transport research focusing on regional/long-distance travel, as well as research that expands beyond urban areas.

Bridging the general and the particular?

As demonstrated in this special issue, it is evident that the emergence of various data sources for studying transport has had a huge impact on transport research, and it has sparked interest in studying mobility related questions from various perspectives. Coming back to the question of ‘generality and particularity’ in transport research, we can see that the articles published in this special issue have produced new and valuable insights, both from the general and the particular. On the one hand, we learnt from large-scale multi-city analyses, for example, by providing evidence on how city size influences the cycling infrastructure development. On the other hand, the majority of the studies in this issue provided learnings from the ‘particular’, that is, case-specific insights considering local contexts, grounded in traditional approaches from (quantitative) transport geography. This collection of papers demonstrates that the path to further strengthening the collaboration between researchers in complexity science and transport geography has started. However, we see that there is a need to continue building bridges between these communities, as there is certainly much to be learned from each other. With this special issue, providing examples from both traditions, we hope to spark interest in collaborating more closely and breaking the silos within which they currently operate.

References

Alessandretti

Natera Orozco

Saberi

, et al. (2023) Multimodal urban mobility and multilayer transport networks. Environment and Planning B: Urban Analytics and City Science 50(8): 2038–2070.

Aras

Ouellette

Jain

(2023) A barrier too far: understanding the role of intersection crossing distance on bicycle rider behavior in Chicago. Environment and Planning B: Urban Analytics and City Science 50(8): 2118–2132.

Beecham

Yang

Tait

, et al. (2023) Connected bikeability in London: which localities are better connected by bike and does this matter? Environment and Planning B: Urban Analytics and City Science 50(8): 2103–2117.

Bruwer

Andersen

(2023) Exploiting COVID-19 related traffic changes to evaluate flow dependency of an FCD-defined congestion measure. Environment and Planning B: Urban Analytics and City Science 50(8): 2220–2237.

Conway

Byrd

van der Linden

(2017) Evidence-based transit and land use sketch planning using interactive accessibility methods on combined schedule and headway-based networks. Transportation Research Record 2653(1): 45–53.

Corcoran

Lewis

(2023) A navigability entropy model for street networks. Environment and Planning B: Urban Analytics and City Science 50(8): 2171–2186.

de Jong

Böcker

Weber

(2023) Road infrastructures, spatial surroundings, and the demand and route choices for cycling: evidence from a GPS-based mode detection study from Oslo, Norway. Environment and Planning B: Urban Analytics and City Science 50(8): 2133–2150.

Derudder

Liu

Hong

, et al. (2019) The shifting position of the journal of transport geography in ‘transport geography research’: a bibliometric analysis. Journal of Transport Geography 81: 102538

Dillman

Czepkiewicz

Heinonen

, et al. (2021) A safe and just space for urban mobility: a framework for sector-based sustainable consumption corridor development. Global Sustainability 4: e28.

10.

Folco

Gauvin

Tizzoni

, et al. (2023) Data-driven micromobility network planning for demand and safety. Environment and Planning B: Urban Analytics and City Science 50(8): 2087–2102.

11.

Franklin

(2022) Quantitative methods I: reckoning with uncertainty. Progress in Human Geography 46(2): 689–697.

12.

Giuliano

Hanson

(2017) The Geography of Urban Transportation. 4th edition. New York: Guilford Publications.

13.

Handy

Niemeier

(1997) Measuring accessibility: an exploration of issues and alternatives. Environment and Planning A 29(7): 1175–1194.

14.

Hanson

(2006). Imagine. Journal of Transport Geography 14(3): 232–233.

15.

Heinen

Mattioli

(2019) Multimodality and CO2 emissions: a relationship moderated by distance. Transportation Research Part D: Transport and Environment 75: 179–196.

16.

Lai

Sun

Schmöcker

J-D

, et al. (2022) Explaining a century of swiss regional development by deep learning and SHAP values. Environment and Planning B: Urban Analytics and City Science 50(8): 2238–2253.

17.

Levinson

(2020) Towards a general theory of access. Journal of Transport and Land Use 13(1): 129–158.

18.

Lowe

(2021) Undone science, funding, and positionality in transportation research. Transport Reviews 41(2): 192–209.

19.

Lunke

Fearnley

Aarhaug

(2023) The Geography of public transport competitiveness in thirteen medium sized cities. Environment and Planning B: Urban Analytics and City Science 50(8): 2071–2086.

20.

Palominos

Smith

(2022) Examining the geometry of streets through accessibility: new insights from streetspace allocation analysis. Environment and Planning B: Urban Analytics and City Science 50(8): 2203–2219.

21.

Poorthuis

Zook

(2023) Moving the 15-minute City beyond the urban core: the role of accessibility and public transport in the Netherlands. Journal of Transport Geography 110: 103629.

22.

Prignano

Font-Pomarol

Morer

, et al. (2023) Infrastructures connecting people: a mechanistic model for terrestrial transportation networks. Environment and Planning B: Urban Analytics and City Science 50(8): 2254–2272.

23.

Reggiani

Verma

Daamen

, et al. (2023) A multi-city study on structural characteristics of bicycle networks. Environment and Planning B: Urban Analytics and City Science 50(8): 2017–2037.

24.

Schwanen

(2017) Geographies of transport II: reconciling the general and the particular. Progress in Human Geography 41(3): 355–364.

25.

Schwanen

(2018) Geographies of transport III: new spatialities of knowledge production? Progress in Human Geography 42(3): 463–472.

26.

Stewart

Byrd

(2023) Half-(head)way there: comparing two methods to account for public transport waiting time in accessibility indicators. Environment and Planning B: Urban Analytics and City Science 50(8): 2187–2202.

27.

Szell

Mimar

Perlman

, et al. (2022) Growing urban bicycle networks”. Scientific Reports 12(1): 6765.

28.

Thompson Sargoni

Manley

(2023) Neighbourhood-level pedestrian navigation using Construal Level Theory. Environment and Planning B: Urban Analytics and City Science 50(8): 2151–2170.

29.

Trope

Liberman

(2012) Construal Level Theory. In: Handbook of Theories of Social Psychology: Volume 1: Sage Publications Ltd. London, pp. 118–134.

30.

von Thünen

(1826) Der isolirte Staat in Beziehung auf Landwirthschaft und Nationalökonomie. Hamburg: Perthes.

31.

Willberg

Tenkanen

Miller

, et al. (2023) Measuring just accessibility within planetary boundaries. Transport Reviews 1–24. doi: 10.1080/01441647.2023.2240958