Urban Currents

Optimal evacuation control in large urban networks with stochastic demand

The study develops a risk-aware Model Predictive Control framework for large-scale vehicular evacuations using the Generalized Bathtub Model to track network-wide trip completion rates. The control policy uses origin gating, such as staged departures or adaptive ramp metering, to minimize a convex objective combining evacuation delay and spatial hazard exposure. Tested on a simulated flood evacuation of Amager Island in Copenhagen using a coupled shallow-water model, the policy reduced the expected area-under-queue by an average of 27% across 10,000 stochastic demand scenarios.

Why it matters — It provides a mathematical proof that the optimal origin-gating profile is monotone decreasing and follows a single-switch bang-bang structure under uniform spatial demand, explaining why early heavy release followed by throttling is mathematically optimal. This theoretical result offers a closed-form seed that significantly simplifies and speeds up numerical optimization for real-time emergency management.

Caveat: The framework's optimal control proof relies on the assumption of a non-decreasing hazard rate in the residual-distance distribution, which may not hold in highly irregular or non-convex evacuation zones.

Transportation Research Part B Methodological · doi

Adequate housing and student achievement: a case of Pakistan during the COVID-19 pandemic

This study analyzed the relationship between housing conditions and academic achievement for 178,610 Pakistani students from early childhood through grade 10. Using multilevel regression on 2020 secondary data from the Annual Status of Education Report (ASER) Pakistan, the analysis measured how housing tenure, structure type, water availability, and internet access relate to student performance during the COVID-19 pandemic.

Why it matters — The research establishes a direct empirical link between physical housing quality, digital connectivity, and educational outcomes in a developing nation context during a major public health crisis, demonstrating that household infrastructure directly shapes a child's capacity to learn when schools are disrupted.

Caveat: The study relies on cross-sectional secondary data collected during the pandemic, which limits the ability to establish strict causal relationships between housing improvements and academic gains.

Journal of Housing and the Built Environment · pdf · doi

Enhancing senior mobility: the role of public transport attributes in improving travel satisfaction and quality of life

This study analyzed survey data from 478 individuals aged 60 and older across five Polish cities to examine how public transport attributes affect travel satisfaction and quality of life. Using Partial Least Squares Structural Equation Modeling (PLS-SEM) and the multidimensional Satisfaction with Travel Scale (STS), the researchers evaluated five travel dimensions: transit stop amenities, service provision, in-vehicle experience, fare affordability, and comfort. The analysis revealed that comfort and the travel environment are the strongest predictors of travel satisfaction, which directly correlates with higher subjective quality of life.

Why it matters — It establishes a direct, empirically measured link between specific, age-friendly transit quality attributes and the overall life satisfaction of older urban residents within a single structural framework. This shifts the focus of senior transit planning from basic functional efficiency to experiential dimensions like vehicle comfort and driver-passenger interactions.

Caveat: The study relies on self-reported survey data collected via online and telephone methods, which may introduce selection bias toward more digitally literate or accessible older adults.

Transportation · pdf · doi

Mapping Hemoglobin Levels Among Women and Children Across Ethiopian Administrative Zones: A Geospatial Small Area Estimation Approach

This study applies geospatial small area estimation to map hemoglobin levels among women and children across Ethiopian administrative zones using survey and census datasets. An area-level Fay-Herriot modeling framework was implemented to compare direct survey estimates, standard Empirical Best Linear Unbiased Predictor (EBLUP), and Spatial Empirical Best Linear Unbiased Predictor (SEBLUP) models. The SEBLUP model achieved the highest precision, reducing the mean coefficient of variation from 13.64 to 6.06 for children and from 13.48 to 6.37 for women.

Why it matters — It demonstrates a reliable method for generating localized health indicators in data-scarce regions where standard survey sample sizes are too small to produce precise administrative zone-level estimates.

Applied Spatial Analysis and Policy · pdf · doi

Scalable machine learning for urban thermal morphology classification: clustering high-resolution microclimate and dynamic mapping of radiant temperature

This study develops an automated machine learning framework to classify urban areas into distinct thermal morphologies using Mean Radiant Temperature (MRT) time series simulated by the LUCIDiT tool. It compares a physics-based heuristic method, a shape-based algorithm (K-Shape), and K-Means clustering applied to both raw time series and engineered thermal features. The K-Means algorithm applied directly to raw time series achieved the highest computational efficiency and mathematical robustness, which was validated by projecting the spatial outputs into a novel 2D Thermal Phase Space.

Why it matters — It provides a scalable way to translate complex, high-resolution microclimate simulation data into intuitive, simplified spatial maps. This allows urban planners and designers without specialized meteorological expertise to identify recurrent microclimatic patterns and target heat mitigation strategies effectively.

Computational Urban Science · pdf · doi

A co-production approach to identify and visualise vulnerability to urban extreme heat in Nottingham, UK

The study co-produced the first Heat Vulnerability Index (HVI) for Nottingham, UK, by combining land surface temperature data with socio-economic indicators like unemployment, disability, and age. Working alongside the Blue-Green Nottingham Group, the researchers mapped the spatial distribution of heat exposure, sensitivity, and adaptive capacity, and used Projection Augmented Relief Models to visualize the results for stakeholders.

Why it matters — It demonstrates how incorporating local stakeholder knowledge and physical 3D visualization tools into index creation produces policy-aligned climate risk maps, moving beyond traditional, purely top-down academic vulnerability assessments.

Caveat: The index relies on land surface temperature as the primary proxy for heat exposure, which may not fully capture the ambient air temperatures experienced by residents.

Sustainable Cities and Society · doi

The role of ports in the transition to low-emission trucks – a multi-system transition analysis

This study analyzes the conditions under which ports can accelerate the adoption of low-emission trucks for hinterland transport. Using a multi-system transition framework, it maps the technological, actor, and institutional connections between hinterland transport and port energy systems under EU and port authority regulations, identifying nine distinct conditions that facilitate this transition.

Why it matters — It establishes that ports are not just transport hubs but pivotal institutional and physical nodes capable of coordinating energy infrastructure and accelerating niche-technology diffusion across previously siloed systems.

Caveat: The findings are based on a literature review rather than primary empirical data or simulations of specific port operations.

Transportation Research Interdisciplinary Perspectives · doi

Hidden patterns of urban mixing across five global cities

This study analyzed travel surveys from five global cities to evaluate how daily mobility influences social mixing. The analysis revealed that individual activity spaces are the primary driver of variations in place exposure, exerting a stronger influence than demographic characteristics or proximity to transit.

Why it matters — It demonstrates that physical proximity to transit and demographic traits are insufficient proxies for actual social exposure, shifting the focus of urban mixing research toward the spatial boundaries of daily activities.

Caveat: The findings rely on self-reported travel surveys, which may underrepresent certain daily trips or informal mobility patterns.

Nature Cities · pdf · doi

Inequities in geographic access to emergency obstetric and newborn care in Greater Accra

This study evaluated geographic accessibility to comprehensive emergency obstetric and newborn care (EmONC) across all 29 districts of the Greater Accra Region using travel time estimates from the Google Maps Directions API. It calculated median travel times and the proportion of reproductive-age women within 15-, 30-, and 60-minute thresholds of both public and private facilities during peak and non-peak traffic. The median travel time to the nearest public EmONC facility was 29.7 minutes during peak traffic, with 90% of women living within a 30-minute travel window.

Why it matters — The analysis demonstrates how traffic congestion and peri-urban location compound socioeconomic disparities, revealing that peak-hour traffic completely cuts off peri-urban women from reaching any public emergency facility within the critical 30-minute window while wealthier urban populations maintain superior access.

Caveat: The study relies on travel time estimates from a proprietary API rather than observed real-world emergency transport times.

Nature Cities · pdf · doi

Identifying Airport and Vulnerable Hinterlands in Urban Agglomerations Using the Plume Model

This study develops a novel method to delineate airport hinterlands by combining airport operational metrics with an adapted plume model. Applying this model to the Yangtze River Delta urban agglomeration, the research maps the service areas of regional airports, identifying PVG, SHA, and HGH as high-intensity point sources whose reach extends along ground transit networks into less competitive zones. It also introduces the concept of vulnerable hinterlands to pinpoint marginal, highly contested boundary areas.

Why it matters — The approach provides a systematic way to map airport competition and identify vulnerable market boundaries, offering planners a tool to prevent over-construction and coordinate aviation resources across dense urban clusters.

Caveat: The empirical validation and mapping are limited to a single urban agglomeration, the Yangtze River Delta.

Journal of Urban Planning and Development · doi

Curated ruins: emotional governance and the memory politics of forgetting-by-design in postcolonial Seoul

This paper analyzes Seoul's postcolonial urban landscape through case studies of the colonial waterworks, the 1995–1996 demolition of the Government-General Building, and the redevelopments of Bukchon Hanok Village and Gwanghwamun Square. It identifies three distinct strategies of 'forgetting-by-design'—infrastructural silence, affective demolition, and aesthetic simulation—to show how the city's spatial planning converts colonial legacies into nationalistic spectacles and emotional comfort rather than critical historical engagement.

Why it matters — It introduces a new conceptual framework for memory governance that demonstrates how urban infrastructure and deliberate spatial design can actively manage public emotion and erase colonial history, shifting the focus of postcolonial urban studies from simple neglect to highly curated, intentional erasure.

Caveat: The study relies on qualitative, historical case studies of high-profile symbolic sites in a single capital city, which may limit its direct application to everyday or non-capital postcolonial urban environments.

City · doi

Passenger occupancy detection and estimation in public transport: a systematic review of technologies and predictive models

This study presents a systematic review of 30 articles on public transport occupancy detection, estimation, and prediction following the PRISMA guidelines. It synthesizes and compares technologies ranging from traditional sensors to advanced multimodal monitoring systems, alongside their corresponding predictive methodologies.

Why it matters — It establishes a comparative baseline of the strengths and limitations of current passenger monitoring technologies, demonstrating how combining multiple data sources improves estimation accuracy for data-driven transit management.

Transportation Letters · doi

A Spatial Clustering Approach of Economic Activities: An Application to the Agribusiness Sector in Brazil's Midwest Region

The study introduces a spatial clustering framework that groups economic activities by the similarity of their geographic distributions, using Wasserstein distance combined with a Shimbel distance matrix to account for spatial connectivity. The method was validated by analyzing 272 agribusiness activities across 466 municipalities in Brazil's Midwest region using data from 2013 to 2021, which successfully classified the industries into four distinct spatial clusters.

Why it matters — It provides a novel empirical tool that integrates geographical distance directly into industry clustering, allowing researchers to identify co-distributed economic activities and spatial patterns that traditional non-spatial clustering methods overlook.

Geographical Analysis · doi

How Do Hurricanes and Federal Aid Affect Eviction Risk? Multi-Year Evidence from Coastal U.S. Counties

Using a panel dataset spanning 2009 to 2018 across coastal U.S. counties, this study analyzed how hurricane frequency and federal disaster aid affect eviction filings and threats. The analysis reveals that hurricanes drive up eviction risks by inflating local rents during the disaster year and for up to two subsequent years. Conversely, higher levels of aggregated federal aid are associated with a reduction in eviction filings and threats four years after a disaster occurs.

Why it matters — The findings establish a direct, multi-year empirical link between natural disasters, rising rents, and renter displacement, while demonstrating that long-term federal aid eventually acts as a stabilizing force to reduce eviction risks.

Caveat: The study relies on county-level aggregated data, which may mask neighborhood-level variations in eviction dynamics and aid distribution.

Journal of Planning Education and Research · doi

"You should see my partners' fingers": A Qualitative Study of Construction Artisans' Perspective on Technical Innovation

This study analyzes the perspectives of 25 construction artisans, including installers, restorers, carpenters, painters, and upholsterers, on technical innovation in North-East Italy using interviews conducted in 2018-2019. Through inductive and deductive thematic analysis, the researchers identified fifteen underlying premises regarding technology adoption, categorized into four core themes: professional identity, financial viability, practical usability, and the role of communication devices.

Why it matters — It provides empirical evidence on how small construction enterprises evaluate technology, revealing that decision-making in these firms is highly integrated because the investor and the end-user are the same person.

Caveat: The findings are based on a small qualitative sample of 25 artisans in a single region of Italy, which may limit their generalizability to larger firms or different cultural contexts.

arXiv · pdf

Global Building Area Estimation Products: How Accurate Are They?

This study evaluates the accuracy of four major global building footprint raster datasets—GHSL, Microsoft's TEMPO, the Global Building Atlas (GBA), and Overture—against a manually labeled ground-truth dataset called ORBITaL-Net. The evaluation measures performance across multiple spatial resolutions, geographic locations, population densities, and income levels. The analysis reveals that GBA and TEMPO achieve the highest overall accuracy, but all four products show a significant drop in accuracy within high-density urban areas and across Africa and Asia.

Why it matters — It establishes the first independent, comparative benchmark for global building area datasets, revealing systematic geographic and density-based biases that users must account for when mapping urbanization or emissions.

Caveat: The evaluation is constrained to the geographic distribution and representation present within the ORBITaL-Net validation dataset.

arXiv · pdf

How Does Urban Context Relate to Residential Building Health? A Vision-POI Fusion Framework for Building-Level Housing Inspection

The study developed a vision-POI fusion framework to assess residential building health across 3,237 buildings in 92 old communities in Qingdao, China, using 25,608 field-acquired images and neighborhood POI data. It aggregates multi-view object detection outputs to the building level and integrates them with POI features extracted within buffers of 500m, 1,000m, and 1,500m using a cost-sensitive Random Forest classifier. Multi-view aggregation improved the building-level Macro-F1 score from 60.84% to 74.95%, while adding POI context yielded a modest further increase to 76.79%.

Why it matters — It demonstrates that while surrounding urban functional context provides a helpful supplementary prior for automated housing inspections, it cannot substitute for direct visual evidence of building defects.

Caveat: The framework was evaluated on a dataset restricted to a single Chinese city, and the performance gains from POI data were found to be modest and highly dependent on the specific category of housing issue.

arXiv · pdf