Urban Currents

Bibliometric analysis and systematic review of spatial injustice and spatial inequality in Sub-Saharan Africa

This paper presents a bibliometric analysis and systematic review of 384 publications on spatial injustice and inequality in Sub-Saharan Africa indexed in the Web of Science between 1981 and 2023. Using VOSviewer, the study maps collaboration networks, publication trends, and thematic clusters, revealing that research is heavily concentrated in South Africa (47.1% of articles), Ghana (12.0%), and Kenya (11.5%).

Why it matters — It maps the geographic and thematic imbalances in African spatial research, establishing that most regional literature is concentrated in just three countries and highlighting underrepresented nations where evidence is lacking to guide inclusive urban planning.

Caveat: The analysis is limited to literature indexed in the Web of Science database, which may exclude relevant local journals, grey literature, or non-English publications from other Sub-Saharan African countries.

Discover Cities · pdf · doi

Environmental readiness and registry competition: A comparative assessment of flag States in maritime decarbonization

This study evaluates the environmental readiness of ten major maritime flag States (Cyprus, Greece, Japan, Liberia, Malta, Marshall Islands, Norway, Singapore, Turkey, and the United States) using a fleet-level framework and harmonized registry data. The framework measures three dimensions: alternative-fuel capability, eco/electronic engine technologies, and SOx scrubber installations. A scenario-based multi-criteria decision analysis reveals that efficiency technologies are widely diffused, scrubbers are mature, and alternative-fuel readiness is highly concentrated, with large open registries maintaining stable top rankings across most policy scenarios.

Why it matters — It establishes a comparative method to measure how prepared different national registries are for maritime decarbonization, revealing a critical governance trade-off between the sheer scale of a registry's fleet and the depth of its technological transition.

Case Studies on Transport Policy · pdf · doi

A multidimensional assessment of housing quality and socio-spatial inequality: Evidence from Ghana

This study constructs a Multidimensional Housing Quality Index (MHQI) using household-level microdata from Ghana's 2021 Population and Housing Census (N = 836,515) to evaluate housing deprivation across nine indicators of structure, space, services, and sanitation. Applying principal component analysis to a tetrachoric correlation matrix, the index reveals a national mean score of 62.2 out of 100, with urban households scoring 73.1 compared to 45.3 for rural households. The Alkire–Foster framework classifies 56.5% of households as housing poor, with district-level scores showing strong spatial clustering (Moran's I > 0.71).

Why it matters — The research provides a transferable, empirical framework to measure multidimensional housing deprivation in the Global South, shifting the policy focus from broad national deficits to targeted, district-level interventions addressing specific local infrastructure gaps.

Habitat International · doi

Exploring associations of neighbourhood greenery, tree canopy, air pollution and walkability with cardiometabolic disease-related hospital admissions at midlife and beyond

This cross-sectional study analyzed 14,378 hospitalized patients aged 45 and older across 697 neighborhoods in Australia's Frankston-Mornington Peninsula region to evaluate how environmental factors relate to cardiometabolic disease. Using multilevel logistic regression on 2021 electronic health records, it found that higher neighborhood greenery (NDVI) and tree canopy cover were associated with lower odds of type-2 diabetes (OR of 0.39 and 0.95 respectively). Conversely, higher exposure to PM2.5 and NO2 was linked to increased odds of both type-2 diabetes and cardiovascular disease, while higher walkability scores were associated with slightly higher odds of cardiovascular disease (OR of 1.04).

Why it matters — It provides empirical evidence of the joint, opposing relationships that local vegetation and traffic-related air pollution have on specific chronic disease admissions within a single suburban-rural fringe population.

Caveat: The study relies on a cross-sectional design and a sample limited to hospitalized patients in a single Australian region, which may not represent the broader population.

Journal of Transport & Health · pdf · doi

International non-binding policies for sustainable land passenger transport in Europe: A scoping review

This scoping review analyzed 5,826 documents from 49 public international databases to identify and categorize over 500 non-binding sustainable land passenger transport policies in Europe published between 2015 and 2024. Using Latent Dirichlet Allocation topic modeling, the study mapped these policies across seven key themes, including transport externalities, governance, and infrastructure, finding that policy volume peaked in 2021 and that position papers, resolutions, and opinions are the most frequent formats.

Why it matters — It establishes that non-binding policies are a highly structured, growing, and diverse instrument in European transport governance, serving as a critical precursor that shapes future binding legislation, especially around inclusive and intersectoral planning.

Transport Policy · doi

Who is the X-minute city for? An E2SFCA assessment of wheelchair accessibility in Zurich City Center

This study evaluated spatial accessibility to public facilities for wheelchair users versus the general population in Zurich's District 1 using the Enhanced Two-Step Floating Catchment Area (E2SFCA) method. It integrated vertical slope data and sidewalk features like curb ramps and surface conditions collected via the participatory ZuriACT project, weighting these barriers based on mobility constraints reported in an online survey of wheelchair users.

Why it matters — It demonstrates that standard proximity-based 'x-minute city' models overestimate accessibility for mobility-impaired residents by failing to account for micro-scale infrastructural barriers, providing a method to quantify these disparities for more inclusive urban planning.

Caveat: The empirical assessment and participatory data collection are limited to a single central district in Zurich.

Journal of Urban Mobility · doi

Decisive factors in sustainable stormwater management

Through a mixed-methods approach, this study traced participant insights across two Swedish cities from 1960 to 2024 using interviews to identify 40 factors shaping stormwater management transitions. These elements were synthesized into 9 key factors and ranked by participants, revealing that external collaboration is perceived as the most critical factor for future sustainable stormwater management.

Why it matters — The study demonstrates that these key decision factors function as dynamic, adaptive levers rather than static barriers or drivers, shifting in influence across time and space to offer a transferable framework that challenges purely technical and economic decision paradigms.

Caveat: The identified factors and their rankings rely on retrospective participant interviews in two Swedish cities, which may reflect localized institutional structures and subjective recall.

npj Urban Sustainability · pdf · doi

Temporal Trends in Freight Train Derailment Rates and Severity under Changing Railway Operations

This study analyzed mainline freight train derailment and traffic data in the United States across two six-year periods, 2012–2017 and 2018–2023, to evaluate how derailment rates and severity changed following the adoption of precision scheduled railroading. Using train length as a proxy for operational shifts, the analysis normalized cause-specific derailment rates by both train-miles and car-miles. The findings show that larger train sizes correlate with increased derailment rates for human factors and train handling, as well as higher derailment severity, though derailments from other causes declined.

Why it matters — It provides empirical evidence linking longer freight train configurations to specific types of derailment risks and increased incident severity, challenging the assumption that long-term safety improvements have continued unabated under modern operating models.

Caveat: The analysis relies on proxy measures for operational changes and is subject to significant limitations in the underlying railroad safety data.

Journal of Transportation Engineering Part A Systems · doi

Caring With/In Place: Future‐Oriented Practices of Care in Local Initiatives in Gothenburg, Sweden

This study analyzes how local community initiatives function as practices of care and future-oriented legacy-making by examining two specific sites in Gothenburg, Sweden: an allotment garden and a neighborhood house. Drawing from a larger study of 35 qualitative interviews on environmental legacy-making, the analysis focuses on two key participants whose active engagements shape these urban spaces. The research employs a constructivist grounded theory approach, combining inductive and abductive data analysis to conceptualize these activities as 'caring with/in place' within a framework of more-than-human assemblages.

Why it matters — It conceptualizes local urban stewardship not merely as environmental footprint reduction, but as an active, relational practice of care that shapes the future trajectories of physical places and fosters attentiveness to non-human elements.

Caveat: The empirical findings are based on an extremely small sample of two in-depth case participants within a single city, which limits the generalizability of the specific behavioral dynamics described.

Area · pdf · doi

Gender homophily and rent premiums in platform-mediated shared housing: Evidence from Ziroom in urban China

Using a dataset of 373,964 bedroom listings over one year in Nanjing, China, this study analyzed how roommate gender composition affects rental prices on Ziroom, the nation's largest shared housing platform. Hedonic regression models controlling for unit, building, and neighborhood features revealed a rent premium for all-female shared units, which is driven not by platform-set base prices but by gendered demand patterns interacting with algorithmic pricing during active leasing and subleasing phases.

Why it matters — The study demonstrates that algorithmic pricing systems on housing platforms can monetize social preferences like gender homophily, translating a desire for same-gender living arrangements into a financial penalty that disproportionately affects female renters.

Caveat: The findings are based on a single platform in a single Chinese metropolis, which may limit generalizability to other cultural contexts or platform structures.

Urban Studies · doi

Automating High-Performance Concrete Mix Design: Deep Learning Framework for Inverse Prediction of Composition from Target Strength and Age

This study developed a multi-tower multilayer perceptron deep learning model to automate high-performance concrete mix design by reversing the traditional prediction task, directly outputting optimal material compositions from target strength and age. Tested against four regression baselines, the model achieved a mean coefficient of determination of 0.739 and an average mean absolute error of 16.712 kg/m3. SHapley Additive exPlanations analysis confirmed the model's predictions align with concrete material science, showing strength dominates cement and admixture predictions while age governs supplementary cementitious materials.

Why it matters — It replaces the conventional, wasteful trial-and-error empirical iterations in concrete design with an automated inverse prediction framework, establishing a capability to design sustainable, high-performance concrete directly from performance objectives.

Transportation Research Record Journal of the Transportation Research Board · doi

Innovative Utilization of Waste Zanthoxylum Seed: Enhancing Rheological and Thermal Stability of Rejuvenated Asphalt Binder

This study evaluates Zanthoxylum seed oil (ZSO), an agricultural waste product, as a rejuvenator and antioxidant for long-term aged Pen60/80 asphalt. Using a Dynamic Shear Rheometer, Bending Beam Rheometer, and Fourier-Transform Infrared Spectroscopy, the research tests binders across simulated aging stages, finding that an optimal ZSO dosage of 8% to 9% by mass restores viscoelastic properties, improves fatigue and low-temperature performance, and reduces the low-temperature performance decay rate to 68% of virgin asphalt.

Why it matters — It demonstrates that an agricultural waste byproduct can serve as a highly effective, anti-aging asphalt rejuvenator, offering a circular economy solution that improves the thermal oxidative stability and cracking resistance of recycled pavements.

Caveat: The findings are based on laboratory-simulated aging and testing of a single control asphalt binder (Pen60/80) rather than field-aged pavement mixtures.

Transportation Research Record Journal of the Transportation Research Board · doi

Becoming-imperceptible: emotional support animals (ESAs) as a form of affective immanence

This study analyzes public comments and interviews with emotional support animal (ESA) owners to critique the framing of these animals' roles as labor. By examining relationships with non-dog ESAs, the research contrasts labor-centric legal definitions with the lived, non-hierarchical dynamics of human-animal companionship.

Why it matters — It challenges the prevailing legal and sociological practice of using 'affective labor' to legitimize ESAs, offering instead a philosophical framework based on Spinozan ethics and Deleuzian 'becoming-animal' to understand human-animal relationships without reducing them to utility or work.

Cultural Geographies · doi

Application of an Interpretable Machine Learning Model to Archaeological Site Prediction: A Case Study of the Three Gorges Region in Chongqing

The study developed an optimized BP neural network model combined with SHAP interpretability to predict archaeological site locations in the Chongqing Three Gorges region. Using Bayesian hyperparameter optimization, the model achieved an AUC of 0.8815 on the test set, representing an improvement of 0.0812 over the baseline. The analysis revealed that sites are distributed in five concentric tiers radiating from the Yangtze River and its tributaries, with elevation, river proximity, NDVI, and land use acting as the primary driving factors.

Why it matters — It provides a highly accurate and interpretable predictive tool that moves beyond black-box machine learning, allowing archaeologists to quantitatively understand the threshold effects and environmental interactions that guided ancient human settlement patterns.

Caveat: The model's predictive rules and environmental thresholds are highly specific to the unique river valley geography of the Chongqing Three Gorges region and may not generalize to different topographic landscapes.

ISPRS International Journal of Geo-Information · doi

Nonuniform and inequitable healthcare accessibility losses during flooding-amplified congestion across US cities

An analysis of 30 of the most populous US cities reveals that peak-hour traffic congestion reduces healthcare accessibility by 4% to 29%, while urban flooding causes an additional reduction of 2% to over 90%. The study models how different flood types, transportation networks, and spatial inequalities interact to restrict access to medical facilities.

Why it matters — It quantifies the compound impact of traffic and flooding on healthcare access across diverse metropolitan areas, demonstrating that coastal flooding disproportionately affects less socially vulnerable populations in the US, while other flood types place unequal burdens on highly vulnerable communities.

arXiv · pdf

Pedestrian Archetypes Extension -- More Pedestrian Models for Autonomous Vehicle Safety Testing

This study expands a behavioral taxonomy for autonomous vehicle safety testing by identifying 7 new pedestrian archetypes from YouTube dash-cam video annotations, building on an existing set of 12 archetypes. Each new archetype is defined by its essential and optional behaviors, distinguished from previous categories, and illustrated with video-frame evidence.

Why it matters — It provides a more comprehensive and realistic set of progressive behavioral patterns for safety simulations, capturing complex real-world pedestrian actions that the original 12-archetype taxonomy could not explain.

Caveat: The new archetypes are derived from a qualitative analysis of YouTube dash-cam videos rather than a systematic, georeferenced traffic dataset.

arXiv · pdf

A Multi-Agent System for 5G Throughput Prediction in Multi-Operator Urban Environments

The researchers developed a Tiered Multi-Agent System (TMAS) that routes edge telemetry to specialized domain micro-agents to predict 5G throughput. The system was validated using a dataset of 48,618 samples collected via drive tests in Sunway City, Malaysia, covering three mobile network operators, three mobility modes, and three traffic profiles. The model achieved a coefficient of determination (R2) of up to 0.931, a Mean Absolute Error as low as 0.53 Mbps, and an agentic routing overhead between 0.004 and 0.126 ms.

Why it matters — It overcomes the generalization limits of monolithic machine learning models in highly variable urban environments, proving that context-aware micro-agents can handle multi-operator and multi-modal network conditions with low enough latency for next-generation wireless orchestration.

Caveat: The system's performance and low latency characteristics were validated on a single dataset from a single master-planned township in Malaysia.

arXiv · pdf