Urban Currents

2026-06-12

7 arXiv categories· 96 journals· 413 candidates — 18 worth your time· 10 without an open abstract

Headline

Graph neural network layer combinations for autonomous vehicle trajectory prediction

Today, together

canon
2 items cite Transport and social exclusion: Where are we now? (Karen Lucas (2012))
coupling
no items share references today
institutions
University of Hong Kong on 3 papers in 30 days

Seven papers carry the tag "Urban Transport and Accessibility", addressing topics such as learning urban access costs, perceived accessibility of public transport, pandemic responses of transport-disadvantaged young adults, travel behavior beyond high-accessibility neighborhoods, sustainable travel in Swedish youth sports, driver risk-taking in Phnom Penh, and active versus car commuting in short trips. Four papers are grouped under "Older Adults Driving Studies", which covers perceived accessibility of public transport, the origins of the U.S. transportation research workforce, motorcycle conflict severity at roundabouts, and commuting choices for short trips. The tag "Traffic and Road Safety" is shared by four papers, focusing on pandemic transport responses, driver risk-taking in Phnom Penh, workplace road safety culture, and motorcycle conflict severity at roundabouts. Three papers carry the tag "Transportation Planning and Optimization", covering urban access costs, perceived accessibility of public transport, and travel behavior beyond high-accessibility neighborhoods. Outside these groups, one paper evaluates nineteen graph neural network layer types for autonomous vehicle trajectory prediction, another analyzes Tehran's urban diplomacy governance model using a survey of ninety municipal officials, and a third uses smartcard data from Perth to analyze transit hub patronage.

preprint

George Daoud, Mohamed El-Darieby
arXiv cs.LG

Autonomous Vehicle Technology and Safety · Traffic control and management · Advanced Graph Neural Networks · graph neural network

A Comparative Study of Graph Neural Network Layer Selection for Interaction Modelling in Driving Trajectory Prediction

This study systematically evaluates 19 different graph neural network (GNN) layer types to determine their effectiveness in modeling spatial and temporal interactions for autonomous vehicle trajectory prediction. The evaluation identifies five top-performing layer combinations, demonstrating that ARMA, Chebyshev, and topology-aware layers consistently outperform other architectures.

Why it matters — It establishes concrete architectural design principles for trajectory prediction models, proving that sum-based aggregation, multi-head attention, and hop-distance weighting yield superior accuracy compared to standard GNN configurations.

Caveat: The findings and recommended layer combinations are constrained to the specific hyperparameter settings explored in the study.

arXiv · pdf

published: Discover Cities

Mohammad Ali Khaliji, Kamran Jafarpour Ghalehteimouri
Discover Cities

Islamic Azad University, Ahvaz Branch

Cross-Border Cooperation and Integration · Global Urban Networks and Dynamics · Public Relations and Crisis Communication · survey · Islamic Azad University, Ahvaz Branch · Kharazmi University

Integrating urban diplomacy and participatory governance in megacities a comprehensive framework for Tehran metropolis in the era of globalization

This study analyzes the governance model of urban diplomacy in Tehran using a survey of 90 municipal officials, including mayors and department vice-presidents. The researchers used Likert-scale questionnaires to measure governance, urban diplomacy, and globalization, validating the relationships with Cronbach's alpha reliability scores of 76%, 83%, and 89% respectively. The analysis identified causal relationships and significant associations between urban diplomacy and governance indicators such as accountability, central consensus, international relations, and management power.

Why it matters — It establishes a empirical link between internal municipal governance structures and a megacity's capacity to conduct international urban diplomacy, identifying specific administrative traits like accountability and consensus-building that facilitate global engagement.

Caveat: The study relies on a relatively small sample of 90 municipal officials within a single city, and its findings are based on self-reported survey data rather than objective measures of diplomatic outcomes.

Discover Cities · pdf · doi

published: Cities

Carol Masingi, Samira Ramezani, Felix Johan Pot, et al.
Cities

Tampere University

Urban Transport and Accessibility · Older Adults Driving Studies · Transportation Planning and Optimization · Tampere University · University of Groningen

Determinants of perceived accessibility of public transport and implications for equity: Evidence, gaps, and a research agenda

This systematic literature review analyzed 67 empirical articles from the Scopus database published between 2004 and 2025 to synthesize how perceived accessibility of public transport is measured and understood. The analysis mapped the relationships between perceived accessibility, transit use, and activity participation, identifying how individual, land-use, transport, and temporal factors are integrated.

Why it matters — The review reveals that existing research frequently isolates individual factors rather than examining their interaction with spatiotemporal constraints, and systematically overlooks intra-group differences in perception among vulnerable populations like young adults and low-income individuals who face transport-related social exclusion.

Caveat: The findings are based on a sample of literature restricted to the Scopus database and note a heavy reliance on quantitative methods, which limits deeper qualitative insights into how transit perceptions are formed.

Cities · doi

published: Journal of Transport & Health

Hitomi Nakanishi, Milica Muminović, Wendi Han, et al.
Journal of Transport & Health

University of Canberra

Urban Transport and Accessibility · Traffic and Road Safety · COVID-19 epidemiological studies · University of Canberra

Behavioural and attitudinal responses to pandemic: a study focusing on young adults with transport disadvantages in low-density built environment

This study analyzed public transport ridership trends in Canberra, Australia, from March 2020 to December 2021, supplemented by qualitative interviews with 20 transport-disadvantaged young adults. Quantitative data revealed a 25% drop in overall transit usage during the pandemic's peak, with a 5% reduction persisting into August 2023, while bus usage remained 8.6% lower in February 2024 compared to 2019. Qualitative analysis identified five key themes, including public transport unreliability, isolation-induced mental health issues, and the role of cars in providing independent mobility.

Why it matters — It provides empirical evidence on how young adults without car access in low-density, car-dependent suburban environments adapt their mobility during and after major disruptions, identifying specific policy interventions needed for different commuter types.

Caveat: The qualitative findings are based on a small sample of 20 participants in a single, low-density Australian city, which may limit generalizability to high-density urban contexts.

Journal of Transport & Health · doi

published: Sustainable Cities and Society

Seung Jae Lieu, Subhrajit Guhathakurta
Sustainable Cities and Society

Georgia Institute of Technology

Urban Transport and Accessibility · Transportation and Mobility Innovations · Transportation Planning and Optimization · Chicago · Georgia Institute of Technology

Why do residents still drive and travel beyond high-accessibility neighborhoods? Examining the challenges to the 15-minute city concept

This study analyzed non-work travel behavior originating from high-accessibility neighborhoods in Chicago to determine why residents still drive or travel outside their 15-minute walkshed. Using a bivariate probit model, it evaluated how destination and mode choices are influenced by individual traits, built environment features, and a newly developed 'harmony of amenities' metric that measures the functional synergy of co-located services based on empirical co-visitation patterns. The findings show that pedestrian-oriented design elements like sidewalks, parks, and compact streetscapes, alongside the functional harmony of local amenities, significantly drive the decision to walk locally rather than drive elsewhere.

Why it matters — It demonstrates that physical proximity to services alone is insufficient to achieve the 15-minute city ideal, proving that the functional relationships between amenities and the quality of the pedestrian environment are critical factors in reducing driving behavior.

Caveat: The empirical analysis is limited to non-work trips originating from high-accessibility neighborhoods within a single city, Chicago.

Sustainable Cities and Society · doi

published: Transportation Research Part A Policy and Practice

Débora Corrêa, Adriano Polpo, Rachel Cardell‐Oliver, et al.
Transportation Research Part A Policy and Practice

The University of Western Australia

Advanced Clustering Algorithms Research · The University of Western Australia

Analysing transit hub patronage: Meta-clustering using smartcard data

The study analyzes public transport demand using 480 million smartcard transactions across 44 transport hubs in Perth, Australia, from 2016 to 2019. It introduces a multi-stage meta-clustering framework that integrates ridership trends, passenger demographics, connectivity, and activity profiles derived from passenger stays. The analysis reveals that hubs serving as 'Generators' and 'Connectors' with park-and-ride or feeder bus infrastructure maintained stable patronage, while 'Attractor' hubs were highly vulnerable to patronage declines.

Why it matters — It provides a method to synthesize multiple distinct data perspectives into a single consensus model, allowing planners to identify how specific land-use functions and infrastructure types dictate a transit hub's resilience to long-term demand shifts.

Caveat: The findings and hub classifications are derived from a single metropolitan transit network, which may limit direct generalization to cities with different spatial structures or transit behaviors.

Transportation Research Part A Policy and Practice · doi

published: Transportation Research Interdisciplinary Perspectives

Miriam Stumpe, David Rößler-von Saß, Natalia Kliewer, et al.
Transportation Research Interdisciplinary Perspectives

Paderborn University

Electric Vehicles and Infrastructure · Transportation and Mobility Innovations · Electric and Hybrid Vehicle Technologies · Paderborn University · Freie Universität Berlin

Impact of vehicle scheduling and strategic transition planning on zero-emission bus systems

The study develops a two-phase optimization framework that first solves the Charging Location and Electric Vehicle Scheduling Problem to create feasible electric bus schedules, and then uses these schedules in a multi-period transition planning model to minimize total cost of ownership. The model determines fleet replacement and charging infrastructure deployment over time, achieving up to an 85% reduction in local emissions compared to a diesel-only baseline.

Why it matters — It demonstrates that planning electric bus transitions using vehicle schedules specifically optimized for electric operations, rather than relying on legacy diesel schedules, significantly lowers the total cost of ownership and accelerates emission reductions by prioritizing long-distance routes with sufficient idle time.

Caveat: The framework's performance and emission reduction metrics are based on experimental simulations rather than a real-world city deployment.

Transportation Research Interdisciplinary Perspectives · doi

published: Transportation Research Interdisciplinary Perspectives

Jisoon Lim, Amirmahdi Tafreshian, Neda Masoud, et al.
Transportation Research Interdisciplinary Perspectives

University of Michigan

Transportation and Mobility Innovations · Human-Automation Interaction and Safety · Digital Transformation in Industry · survey · University of Michigan · Michigan Department of Transportation

Enhancing transportation education to reflect technological advancements in connected and automated vehicles

This study designed a workshop and distributed a pre-workshop survey to academic and industry experts to identify workforce development needs and educational obstacles in the field of connected and automated vehicles (CAVs). The investigation structured its findings around four core themes: next-generation infrastructure, human factors, modeling and simulation, and travel behavior.

Why it matters — It shifts the academic focus from purely researching CAV impacts to actively preparing the future transportation workforce, establishing a structured agenda for updating university curricula to match rapid private-sector technological advancements.

Caveat: The findings are based on qualitative expert consensus from a single workshop and survey, rather than a quantitative assessment of labor market demand or curriculum efficacy.

Transportation Research Interdisciplinary Perspectives · doi

published: Transportation Research Interdisciplinary Perspectives

Andrew J. Bae
Transportation Research Interdisciplinary Perspectives

Carnegie Mellon University

Research, Science, and Academia · Older Adults Driving Studies · Conservation, Ecology, Wildlife Education · United States · China · Carnegie Mellon University

Global sources of the U.S. transportation research workforce

This study analyzes a newly assembled roster of 553 transportation-focused faculty members across 151 ranked civil and environmental engineering departments in the United States to trace their undergraduate origins. The analysis reveals that while domestic-trained graduates represent the largest single group at approximately one-third of the workforce, they are not a majority, with China serving as the second-largest source at just over one-fifth. Longitudinal cohort analysis shows a declining share of U.S.-trained faculty and an increasing share of Chinese-trained faculty since the 1990s.

Why it matters — It establishes the first empirical, field-level baseline of the geographic origins of the U.S. transportation academic workforce, demonstrating that the country's research capacity is fundamentally dependent on global higher education and international talent mobility.

Caveat: The study's scope is limited to faculty within civil and environmental engineering departments, which may exclude transportation researchers working in other academic disciplines like urban planning, geography, or economics.

Transportation Research Interdisciplinary Perspectives · doi

published: Transportation Research Interdisciplinary Perspectives

Dionysios Polemis, Konstantinos Likas
Transportation Research Interdisciplinary Perspectives

University of Piraeus

Economic Sanctions and International Relations · Maritime Navigation and Safety · Maritime Security and History · University of Piraeus

Characteristics of the shadow fleet phenomenon in crude oil shipping: A quantitative analysis of vessel characteristics, governance attributes, and sanctions-risk classification

This study analyzes 654 crude-oil tankers classified by S&P Global Market Intelligence using binary GLM, ordered logistic, and ridge-penalised regression to identify the determinants of sanctions-risk banding. The models, which achieved a binary GLM AUC of 0.965 and a ridge logistic cross-validated AUC of 0.937, reveal that flag state and ultimate beneficial owner jurisdiction are the primary predictors of risk, while physical vessel size loses explanatory power when these governance controls are added.

Why it matters — It demonstrates that maritime sanctions compliance should pivot from traditional size-anchored screening instruments to governance-based frameworks, proving that regulatory and ownership attributes are far more predictive of shadow fleet participation than physical ship characteristics.

Caveat: The analysis relies on a specific classification of 654 tankers by a single commercial provider, S&P Global Market Intelligence, which may not capture the entirety of the global shadow fleet.

Transportation Research Interdisciplinary Perspectives · doi

published: Transportation Research Interdisciplinary Perspectives

Dmitriy Muzylyov, О. Pavlenko, Vitalii Ivanov, et al.
Transportation Research Interdisciplinary Perspectives

Kharkiv National Automobile and Highway University

Optimization and Mathematical Programming · Supply Chain and Inventory Management · Cognitive Science and Mapping · Kharkiv National Automobile and Highway University · Technical University of Košice · Sumy State University

Modeling agricultural supply chain operations with a petri net-based decision support system: a case study of agribusiness

This study develops a Petri Net-based decision-support model to simulate agricultural supply chain operations under disruptive conditions in Ukraine. Tested on an agribusiness case study, the model defines five operational threshold regimes and evaluates fleet-sizing configurations for 13-hour and 16-hour service level agreements. Empirical validation over 10 campaign days demonstrated high predictive accuracy, yielding a Mean Absolute Error of 0.44 hours and a Mean Absolute Percentage Error of 2.73%.

Why it matters — It provides agricultural enterprises with a quantitative tool to balance harvesting and transport capacities, preventing 'transport-collapse' bottlenecks where adding harvesting equipment without sufficient trucking actually increases shipping times from 11.8 to over 19 hours.

Caveat: The empirical validation of the model is limited to a single agribusiness case study over a short 10-day campaign period.

Transportation Research Interdisciplinary Perspectives · doi

published: Transportation Research Interdisciplinary Perspectives

Ellen Lagrell, Anders Larsson
Transportation Research Interdisciplinary Perspectives

University of Gothenburg

Sport and Mega-Event Impacts · Urban Transport and Accessibility · Recreation, Leisure, Wilderness Management · interview · University of Gothenburg

(Un)sustainable travel and organised sports in Sweden: A social practice approach

This study analyzes the travel patterns of youth and children participating in non-elite soccer, jujutsu, horse riding, and orienteering in Sweden. Using a practice-theoretical framework applied to 33 interviews, it examines how the spatial and temporal organization of these sports, including equipment needs, facility functions, and competition structures, drives car dependency.

Why it matters — It identifies specific structural interventions to reduce car use, such as using sports facilities for equipment storage, adjusting training schedules, and leveraging team structures for collective travel coordination, rather than treating sports travel as an unchangeable aspect of family life.

Caveat: The findings are based on a qualitative sample of 33 interviews, which may not represent the full diversity of youth sports travel across different socio-economic or geographic regions in Sweden.

Transportation Research Interdisciplinary Perspectives · doi

published: Transportation Research Interdisciplinary Perspectives

Jose Rafael Martinez Garcia, Chheng Kimlong, Alfredo C. Panizales, et al.
Transportation Research Interdisciplinary Perspectives

National University of Singapore

Traffic and Road Safety · Urban Transport and Accessibility · Crime, Deviance, and Social Control · regression · National University of Singapore

Risk-taking, informal norms, and social identity in urban traffic: A mixed-methods study of driver behaviour in Phnom Penh, Cambodia

This study analyzed driver risk-taking in Phnom Penh, Cambodia, using a mixed-methods approach that combined a 29-item behavioral questionnaire administered to 150 drivers and motorcyclists with eight semi-structured interviews with local officials and practitioners. Exploratory factor analysis identified five behavioral dimensions to construct a composite risk-taking index, while logistic regression models evaluated how demographic traits and compliance relate to driving habits. The quantitative analysis revealed few statistically significant differences in average risk-taking across demographic groups, but qualitative interviews showed that younger and male drivers justify risky maneuvers through distinct identity-based and peer-group norms.

Why it matters — The study demonstrates that traffic risk in rapidly urbanizing contexts is a socially embedded and interactionally negotiated phenomenon rather than just an individual behavioral deficit. This shifts the focus of road safety planning from purely infrastructural and enforcement-based solutions toward interventions that target informal social norms and peer dynamics.

Caveat: The findings rely on a relatively small sample size of 150 survey respondents and eight interviewees within a single city, which may limit the generalizability of the quantitative risk-taking index.

Transportation Research Interdisciplinary Perspectives · doi

published: Transportation Research Interdisciplinary Perspectives

Hayley McDonald, Angela Batson, David Logan, et al.
Transportation Research Interdisciplinary Perspectives

Monash University

Occupational Health and Safety Research · Traffic and Road Safety · Safety Warnings and Signage · Monash University · Government of Victoria

Workplace road safety: How does safety culture and safety climate impact work-related driving?

This scoping review analyzed 7 peer-reviewed papers and 38 grey literature resources, selected from an initial pool of 170 documents, to examine how safety culture and safety climate affect work-related driving.

Why it matters — The review reveals a disconnect in workplace road safety responsibility: while grey literature emphasizes that managers and leaders hold legislative duties, peer-reviewed academic studies tend to focus disproportionately on individual driver behavior rather than leadership's role in fostering a safe culture.

Caveat: The findings are constrained by a very small volume of academic literature meeting the inclusion criteria, limiting the depth of the peer-reviewed evidence base.

Transportation Research Interdisciplinary Perspectives · doi

published: Transportation Research Interdisciplinary Perspectives

Pongpiwat Chanwongsa, Thaned Satiennam, Wichuda Satiennam
Transportation Research Interdisciplinary Perspectives

Khon Kaen University

Traffic and Road Safety · Traffic control and management · Older Adults Driving Studies · Khon Kaen University

Motorcycle conflict severity at roundabout entries: the roles of rider behavior and self-enforcing geometry

This study analyzed 950 motorcycle entry interactions captured via unmanned aerial vehicles at four single-lane roundabouts in Khon Kaen University, Thailand. Using Post-Encroachment Time to measure surrogate safety, the researchers classified conflicts into four severity levels and used an ordered probit model to evaluate the influence of rider behavior, traffic conditions, and roundabout geometry. The analysis revealed that yielding behavior had the strongest impact on conflict severity, and that among geometric factors, only the radius of deflection significantly influenced safety outcomes.

Why it matters — It demonstrates that path-based geometric measures, specifically the radius of deflection, are more reliable indicators of motorcycle safety than traditional roundabout dimensions, showing that geometric design influences safety primarily by forcing riders to adjust their speed and trajectory.

Caveat: The findings are based on a small sample of four roundabouts within a single university campus, which may limit generalizability to broader urban or highway environments.

Transportation Research Interdisciplinary Perspectives · doi

published: Transportation Research Interdisciplinary Perspectives

M.C. Melin, J. Tuominen, P. Salo
Transportation Research Interdisciplinary Perspectives

University of Turku

Urban Transport and Accessibility · Travel-related health issues · Older Adults Driving Studies · University of Turku

Correlates of active or car commuting in short trips: associations of health and climate concern

This study analyzed survey data from 534 full-time Finnish workers with commutes of 10 kilometers or less to identify factors influencing active and car commuting. Using multinomial logistic regression, it evaluated the association of 27 variables—including climate concern, social norms, and subjective barriers—with commuting habits categorized as never, sometimes, or always. The analysis revealed that health-related motivation consistently distinguished habitual active commuters, while climate concern was primarily associated with reduced car use rather than increased walking or cycling.

Why it matters — It demonstrates that the motivators for reducing driving are distinct from those that encourage active travel, suggesting that environmental messaging and health-focused campaigns target different segments of commuter behavior.

Caveat: The findings are based on self-reported, cross-sectional survey data, which limits the ability to establish causal relationships between motivations and commuting choices.

Transportation Research Interdisciplinary Perspectives · doi

preprint

Paula Joy B. Martinez
arXiv cs.LG

Urban Transport and Accessibility · Transportation Planning and Optimization · Smart Parking Systems Research

Learning Urban Access Costs from Origin-Destination Flows via Inverse Optimal Transport

The study introduces an inverse optimal transport framework to estimate latent urban access costs from origin-destination data, applying it to school choice in the Philippines. Using two models—an interpretable distance-banded model with a subsidy term and a neural cost model trained via a differentiable Sinkhorn pass—the researchers analyzed 283,016 learner trips across 23,820 observed flows in the country's most populated region. The framework successfully quantifies a subsidy-equivalent distance, measuring the exact kilometers of perceived travel cost offset by the national private school subsidy.

Why it matters — It provides a method to convert standard administrative origin-destination flows into interpretable planning metrics, allowing planners to measure how households trade off distance against financial subsidies when accessing public-private service networks.

Caveat: The empirical application is limited to a single sector and region, focusing specifically on school enrollment flows within the most populated region of the Philippines.

arXiv · pdf

preprint

Ariton Verush, Vaibhav Motwani
arXiv cs.CY

Crime Patterns and Interventions · Cybercrime and Law Enforcement Studies · Digital and Cyber Forensics

A Fuzzy Logic Framework for Community-Aware Crime Hotspot Detection: Prototype Application Architecture and Exploratory Validation for an Urban Computing Platform

This study presents a fuzzy logic-based framework and system architecture for real-time urban crime hotspot detection and notification. The platform integrates citizen reports, historical crime data, and contextual urban indicators to model partial risk and uncertainty rather than binary classifications. An exploratory validation of the system was conducted with 25 participants to evaluate usability, trust, privacy concerns, and perceived usefulness.

Why it matters — It demonstrates how fuzzy logic can represent gradual, context-dependent urban risks and incomplete information, offering a community-aware alternative to traditional, reactive law enforcement reporting systems.

Caveat: The system's evaluation is limited to user perceptions of usability and trust among a small group of 25 participants, and does not provide evidence of actual crime reduction or real-world predictive accuracy.

arXiv · pdf

Still cited

Karen Lucas (2012), Transport and social exclusion: Where are we now? 2 of today's items cite it · 59 of 4454 in the archive stand on it

Also published today

These appeared today in journals we track. Their abstracts are not openly available, so we cannot summarise them.