Urban Currents

PLAYING WITH PREFIGURATION: Ludic Spaces and Prefigurative Organizing for Children's Right to the Mobile City

This paper conceptualizes child mobility activism—specifically mobility collectives, street experiments, and demonstrations—as ludic spaces that prefigure children's right to the mobile city. It analyzes how activists use these play-based spaces, which rely on real emotional stakes and disruptive rhythms, as a foundation for prefigurative organizing.

Why it matters — It connects the political theory of play with prefigurative politics, demonstrating how temporary, playful occupations of street space can serve as real-world organizing tools for children's urban mobility rights.

Caveat: The analysis is conceptual and notes that these prefigurative organizing processes are highly contextual and fragmented.

International Journal of Urban and Regional Research · doi

Geographic inequalities in cardiologist distribution and cardiovascular disease mortality in the Greek National Health System: a geo-epidemiological study

This study mapped the geographic distribution of public-sector cardiologists and cardiovascular disease (CVD) mortality across Greece's prefectures using 2022 population-based mortality data and national professional registries. Prefectures were categorized by comparing their age-standardized mortality ratios against the national average alongside their cardiologist density per 100,000 inhabitants. The analysis revealed significant spatial mismatches, with several prefectures exhibiting high CVD mortality rates despite having lower-than-expected cardiologist availability.

Why it matters — It demonstrates that public healthcare resources in Greece are not aligned with regional health needs, showing that lower cardiologist supply systematically coincides with higher mortality in areas characterized by lower tertiary education levels, lower urbanization, and insularity.

Caveat: The study uses an ecological design at the prefecture level, which prevents drawing direct causal links between individual cardiologist access and personal health outcomes.

GeoJournal · pdf · doi

A Model Proposal for Assessment Housing Quality in Urban Regeneration with ANP: The Case of Kadıköy, İstanbul

The study develops an integrated housing quality assessment model for urban regeneration projects by combining international standards with local context. It uses the Analytical Network Process (ANP) to map and weight relationships between qualitative criteria based on expert opinions, and validates the model using survey responses from property owners in the Kadık%oy District of İstanbul.

Why it matters — It establishes a structured, quantitative framework to measure qualitative housing standards under T%urkiye's disaster risk regeneration laws, filling a regulatory gap where explicit quality criteria for rebuilt residential settlements were previously undefined.

Caveat: The empirical validation of the model's weights relies on survey data from a single district, Kadık%oy, which may limit its direct applicability to other regions with different socio-economic profiles.

Journal of Housing and the Built Environment · pdf · doi

Revisiting the political economy of housing in Ghana: progress or stagnation

This study analyzes Ghana's housing crisis from 1963 to 2023 using qualitative policy analysis of documents from various state agencies. It traces how neoliberal policies shifted the state's role from housing provider to market facilitator, identifying structural barriers such as fractured institutional arrangements, high interest rates, limited mortgage access, and political interference in funding.

Why it matters — It demonstrates that Ghana's persistent housing shortages and resulting social inequalities are structural outcomes of neoliberal policy frameworks rather than mere planning failures, indicating that resolving the crisis requires a fundamental shift toward state-financed housing and tenant protections.

Caveat: The analysis relies on qualitative policy documents and agency data without incorporating primary empirical data from affected residents or housing developers.

Journal of Housing and the Built Environment · pdf · doi

Public Perceptions of Urban Inclusivity from a Disability-Friendly Perspective: Attention vs. Satisfaction

This study evaluates urban inclusivity from a disability-friendly perspective across 59 Chinese cities by analyzing public attention via Baidu Search Index data and public satisfaction through survey data. The analysis reveals significant spatial disparities, with satisfaction scoring higher in eastern and megacity regions, while public attention is more widely dispersed and strongly linked to internet penetration, education, and healthcare access.

Why it matters — It establishes that public concern for disability-inclusive infrastructure does not automatically align with perceived governance outcomes, revealing critical attention-satisfaction mismatches that can help planners target interventions in cities with high public demand but low satisfaction.

Caveat: The findings rely on digital search traces as a proxy for public attention, which may skew representation toward internet-active populations.

Applied Spatial Analysis and Policy · pdf · doi

Urban expansion driven by coexisting diffusion and coalescence processes in a rapidly growing Sub-Saharan city

This study analyzed the spatial expansion of Abomey-Calavi, Benin, between 1988 and 2023 using Landsat imagery classified with a supervised maximum likelihood algorithm. Built-up areas expanded tenfold from 2,213 to 23,601 hectares, primarily replacing vegetation and agricultural land which shrank by 45%. Landscape metrics and patch analysis revealed a mean annual urban growth rate of 6.11%, characterized by urban coalescence within 3 km of the city core and active diffusion beyond 6 km.

Why it matters — The findings demonstrate that urban diffusion and coalescence can occur simultaneously rather than sequentially, challenging the dominant global model of urban growth stages and providing empirical evidence from a rapidly growing Sub-Saharan secondary city.

Caveat: The analysis relies on a single case study of a secondary city experiencing spillover from a nearby metropolis, which may limit generalizability to other secondary cities lacking similar metropolitan proximity.

Discover Cities · pdf · doi

Assessing and mapping public visual perception across urban public space typologies based on geotagged social media images

This study developed a deep learning framework using ResNet-50, Mask2Former, and GIS spatial analytics to analyze public visual preferences across four public space typologies using geotagged Flickr images. Focusing on a 2 km radius around Sydney's Darling Harbour, the model classified images with high expert consensus (Cohen's kappa up to 0.92) and found that while waterfronts attracted 52% of visual engagement, the highest positive perceptions clustered in hybrid nodes where blue, green, and built environments overlap. Tumbalong Park outperformed all single-typology environments across six perception dimensions, showing a Cohen's d of +0.48 for 'Beautiful' and -0.81 for 'Depressing'.

Why it matters — It demonstrates that positive urban experiences are driven by the spatial synergy of integrated blue, green, and built elements rather than water frontage or greening alone, providing a scalable workflow to guide human-centered urban design.

Caveat: The findings are based on a single, highly developed 2 km study area around Sydney's Darling Harbour, which may limit generalizability to less dense or non-waterfront urban contexts.

Applied Geography · doi

Does road investment reduce average travel time and travel time variability?: Empirical evidence from Japan

This study evaluated the impact of a new bypass parallel to a national highway in a local Japanese city using GPS-based journey records of 141 million trips before and 167 million trips after the opening. Using propensity score matching across 18 data-screening scenarios, the analysis found that the bypass reduced average route-level travel times by 1.02 to 2.75 minutes and planning times by 0.55 to 3.21 minutes. However, the effects on travel time variability, measured via standard deviation and buffer time, were not robust across the different scenarios.

Why it matters — It provides empirical, GPS-derived evidence that road investments can reliably reduce average travel times and planning times, while demonstrating that measured improvements in travel time variability are highly sensitive to how temporal data is screened.

Caveat: The findings are based on a single bypass project in one local Japanese city, which may limit generalizability to other urban contexts or larger network expansions.

Research in Transportation Economics · doi

The hidden cost of fast delivery: Consumer willingness to trade delivery time and cost for food delivery rider safety

A stated preference experiment surveyed 350 regular food delivery users in Amsterdam and Copenhagen to evaluate how safety labels indicating rider crash risk affect consumer choices. Using a hybrid latent class choice model, the study analyzed consumer willingness to accept longer wait times or higher costs in exchange for safer working conditions for riders.

Why it matters — It demonstrates that a distinct segment of urban consumers is willing to trade personal convenience and money for rider safety when presented with explicit risk labels, offering a non-regulatory pathway to mitigate the traffic hazards faced by gig workers.

Caveat: The findings rely on stated preference survey data from two specific European cities, which may reflect higher baseline safety awareness than the actual behavior of the broader consumer population.

Transport Policy · doi

Mitigating network traffic congestion via link- and path-based incentives

This study formulates and compares link- and path-based incentive schemes designed to reduce urban traffic congestion under budget constraints using non-linear optimization with complementarity constraints. The researchers mathematically proved that the feasible region of the link-based model is a subset of the path-based model, and developed a column generation-based iterative algorithm to solve both models. Numerical experiments across various transport networks were conducted to evaluate the algorithm's scalability and the models' performance in reducing total travel time.

Why it matters — It mathematically establishes that path-based incentives are inherently more effective than link-based incentives at shifting user equilibrium toward a system optimum under the same budget, providing a theoretical foundation for designing more efficient credit or subsidy programs in urban traffic management.

Caveat: The performance and superiority of the schemes are demonstrated through mathematical proofs and numerical experiments on simulated transport networks rather than real-world empirical deployments.

Transportation Research Part B Methodological · doi

Critical success factors for IoT-based curbside management systems

The study analyzes 44 international projects and expert interviews to identify nine design dimensions and 26 configurations for IoT-based curbside management. It establishes five critical success factors categorized under information quality (accurate parking durations, accurate parking positions, and information completeness) and system quality (usability/flexibility and privacy).

Why it matters — It provides a structured framework that addresses why previous IoT curbside pilots failed, shifting the focus from purely technical capabilities to balancing enforcement needs with user buy-in and stakeholder conflicts.

Caveat: The findings are based on a qualitative morphological analysis of past projects and expert interviews rather than empirical performance metrics of deployed systems.

Transportation Research Interdisciplinary Perspectives · doi

Integrating e-bikes and bikesharing with public transport: Impacts on job accessibility and equity

The study models job accessibility and spatial equity in Sao Paulo, Brazil, by integrating e-bikes and bikesharing with public transport across access-only, egress-only, and fully integrated scenarios. Cycling travel times were calculated using GPS data from the Tembici bikesharing system to compare e-bikes, conventional bicycles, and walking. The analysis reveals that while e-bikes and bikesharing at egress yield the largest accessibility gains, particularly for lower- and middle-income groups at longer travel times, absolute accessibility remains concentrated in wealthy central areas.

Why it matters — It quantifies the specific accessibility and equity trade-offs of different multimodal integration strategies, demonstrating that while cycling-transit integration reduces income-based and spatial inequalities, it cannot fully overcome deep-seated peripheral deficits without expanding physical station networks and infrastructure.

Caveat: The findings are constrained by the existing, uneven spatial distribution of the Tembici bikesharing stations and cycling infrastructure in Sao Paulo.

Journal of Urban Mobility · doi

Investigating Whether Autonomous Shuttles Are Ready to Provide Regular Transit Service: A Review and Expert Interviews

This study evaluates the readiness of autonomous shuttles for regular transit service by combining a comprehensive literature review of pilot deployments with insights from eleven interviews with experts across academia, industry, and public and private agencies. The evaluation assesses these systems across technological, infrastructural, regulatory, and public acceptance dimensions.

Why it matters — It establishes that autonomous shuttles are not yet ready for full-scale transit integration due to unresolved safety issues in mixed traffic, vague cybersecurity policies, high sensor data costs, and funding constraints, moving beyond optimistic industry projections to identify specific systemic barriers.

Caveat: The findings rely on a relatively small sample of eleven expert interviews to supplement the literature review.

Journal of Transportation Engineering Part A Systems · doi

A Spatial Analysis of Urban Space Albedo According to Urban Surfaces Color

The researchers developed a Python 3.10 program that estimates urban surface albedo by extracting dominant colors from street-view images and mapping them to reflectance values. Applying this tool to the densely built Kowloon District in Hong Kong, they mapped spatial variations in albedo and identified lower albedo values in the southern parts of the district due to a concentration of dark-colored surfaces.

Why it matters — It provides a highly scalable, low-cost method for mapping urban albedo across complex street canyons using existing street-level imagery, replacing expensive and localized physical measurement campaigns.

Caveat: The method relies on dominant image colors as a proxy for actual material reflectance, which may introduce inaccuracies compared to direct physical measurements.

Journal of Urban Planning and Development · doi

The Purposes, Possibilities, and Potentials of Research in Geography Education: International Perspectives on Research Directions

This paper synthesizes international perspectives on the current challenges, purposes, and future directions of research in geography education.

Why it matters — It establishes a shared agenda for a niche academic field to help researchers coordinate their efforts and address systemic institutional challenges.

Journal of Geography · pdf · doi

Towards Inclusive Mobility Modeling: Characterizing and Evaluating Elderly Trajectory Patterns in Urban Systems

Using the 2016-2020 Jersey City Citi Bike dataset, this study analyzed the distinct mobility signatures of elderly riders compared to younger ones, finding they have smaller activity spaces (958 m vs. 1,189 m), lower mobility entropy (1.82 vs. 4.15), and asymmetric off-peak travel times. The researchers evaluated how these differences affect synthetic trajectory generation by training a first-order Markov chain and a fine-tuned Qwen3-4B model on different demographic subsets. Models trained on the majority-dominated full population systematically misrepresented elderly behavior, with the Markov model overestimating elderly step length by 4.5% and dwell time by 8.9%.

Why it matters — It demonstrates that training mobility models on majority-dominated datasets systematically biases synthetic trajectory generation against underrepresented subgroups, and proves that using more advanced models like Large Language Models does not resolve this fidelity gap when demographic representation is lacking.

Caveat: The empirical analysis is limited to bike-sharing trajectory data from a single mid-sized city.

arXiv · pdf

When Shortest Isn't Safest: A Design Science Approach to Senior-Friendly Pedestrian Routing

The researchers developed a senior-friendly pedestrian routing engine using an A* algorithm on an OpenStreetMap network enriched with height data, benches, toilets, and shelters. The routing system was designed based on 11 semi-structured interviews and evaluated through a field-based walking study with 14 older adults who compared the custom routes against standard baselines.

Why it matters — It demonstrates that optimizing pedestrian routing for age-friendly infrastructure and barrier avoidance, rather than just shortest distance or time, directly aligns with the actual preferences and mobility constraints of older walkers.

Caveat: The routing engine and design principles were evaluated with a small sample of 14 older adults in a single field-based study, which may not capture the full diversity of late-life mobility needs.

arXiv · pdf · doi

Beyond Demographics: BIM Engagement and Job Satisfaction Among AEC Professionals, A Machine Learning Pilot Study

This pilot study analyzed survey responses from 104 architecture, engineering, and construction (AEC) professionals using Spearman rank correlations, logistic regression, and Classification and Regression Tree (CART) modeling to evaluate predictors of job satisfaction. The analysis utilized a 27-item Job Satisfaction Index and found that the proportion of project work completed using Building Information Modeling (BIM) was the only significant predictor of job satisfaction. Demographic factors, including age, gender, education level, and professional experience, showed no significant relationship with satisfaction levels.

Why it matters — The study demonstrates that technological engagement and the integration of BIM tools in daily workflows are more influential in shaping employee satisfaction than traditional demographic characteristics, offering a new lever for firms looking to improve retention and workplace well-being.

Caveat: The findings are based on a small pilot study of 104 survey participants, which may limit the generalizability of the predictive models.

arXiv · pdf

Intelligent Base Station Deployment in Urban Wireless Networks: A Geographic Data-Informed Digital Twin Approach

The researchers developed an intelligent base station deployment framework that combines a geographic data-informed digital twin with deep reinforcement learning to optimize macro base station placement. The digital twin uses a sample-free radio map prediction model and a diffusion-based generative model for trajectory synthesis to estimate kilometer-scale signal strength in milliseconds. Tested in real-world urban scenarios, the framework achieved up to 98.9% of idealized benchmark performance while reducing optimization overhead by more than 99%.

Why it matters — It enables network operators to optimize large-scale base station placement using only open geographic data, eliminating the need for expensive on-site measurements, real user trajectories, or computationally heavy ray-tracing simulations.

arXiv · pdf

Lost in the Tail: Addressing Geographic Imbalance in Urban Visual Place Recognition

The researchers developed Distribution-Aware Place Recognition (DAPR), a plug-in framework that corrects geographic bias in visual place recognition models by rebalancing gradient contributions between frequently and rarely photographed locations. Evaluated on the large-scale SF-XL benchmark, the framework improved baseline classification-retrieval performance by 18.3% on test set v1 and 6.7% on test set v2, while demonstrating consistent accuracy gains when integrated into existing models tested on the MSLS and Pitts30k datasets.

Why it matters — It mitigates a systemic bias where location-matching algorithms fail in sparsely photographed urban areas because they are trained on datasets heavily skewed toward popular landmarks.

arXiv · pdf