Urban Currents

Your place or mine? Overcoming neoliberal placemaking as commodification of built environments in the inclusive city

This paper critiques neoliberal placemaking practices through a case study of Rotterdam's Urban Leisure and Tourism Living Lab. It analyzes how top-down regeneration initiatives commodify urban spaces, leading to gentrification and displacement, and contrasts these with citizen perspectives from local urban activation projects.

Why it matters — It demonstrates how well-intentioned, participatory-labeled planning initiatives can actively marginalize vulnerable communities, offering a framework for bottom-up, community-focused placemaking that aligns with the paradigm of New Urban Tourism.

Caveat: The empirical evidence is drawn from a single case study of a specific tourism and leisure living lab in Rotterdam, which may limit direct generalizability to other urban planning contexts.

Cities · doi

When transit stops: Causal evidence on travel times from a transit strike in Quebec city

This study measures the impact of a 2025 public transit strike on travel times during a major summer music festival in Québec City, Canada. Using smartphone tracking data and a difference-in-differences framework, it compares travel times to and from the festival site between 2023 and 2025. The analysis reveals that trips were significantly longer when transit services were paralyzed, with the most severe delays concentrated among travelers originating from the city center and inner suburbs.

Why it matters — It provides direct causal evidence of the travel-time savings generated by public transit, demonstrating that transit systems yield meaningful economic benefits even in medium-sized cities with low transit modal shares.

Caveat: The findings are based on travel patterns associated with a highly localized, high-demand event (a music festival) and may not fully reflect everyday commuting conditions during a strike.

Cities · pdf · doi

How noise exposure shapes resident preferences for UAM-friendly neighborhoods: Evidence from Guangzhou, China

This study analyzed resident preferences for Urban Air Mobility (UAM) neighborhood integration using stated preference survey data and a mixed logit model in Guangzhou, China. It measured how choices are influenced by drone docking pad installation costs, vertiport proximity, social influence, property management fees, and noise exposure. The analysis reveals that younger, wealthier, and tech-oriented residents favor UAM integration, while highly educated individuals and those living on floors 1 to 7 show the highest sensitivity to noise.

Why it matters — It provides quantitative evidence on how residents trade off the convenience of aerial transit infrastructure against localized noise pollution, establishing which demographic groups and building levels require targeted mitigation during neighborhood renewal.

Caveat: The findings are based on stated preference survey data rather than observed behavior in an active Urban Air Mobility deployment.

Case Studies on Transport Policy · pdf · doi

How neighborhood crime, built environment, life events, and attitudes influence vehicle acquisitions: A retrospective analysis of purchase, body type, and vintage decisions

This study analyzed how neighborhood crime, built environment, and life events influence household vehicle purchases, body types, and vintage choices using a 10-year retrospective survey from Chicago. A nested logit model was used to link household decisions to violent and property crime data, socio-demographics, and transit access. The analysis found that exposure to crime increases the likelihood of purchasing a vehicle and shifts preferences specifically toward used vehicles, while higher residential density and transit access decrease purchase probability.

Why it matters — The study establishes a direct empirical link between neighborhood crime exposure and vehicle ownership decisions, demonstrating that personal safety concerns drive car dependency even when transit and density are present.

Caveat: The findings are based on retrospective survey data from a single city, Chicago, which may introduce recall bias and limit generalizability to other urban contexts.

Case Studies on Transport Policy · pdf · doi

Transit-induced household migration: A survival analysis of heterogeneous income effects near urban rail corridors

This study analyzed household migration patterns near Charlotte's LYNX Blue Line and its Extension using a 16-year panel of 437,606 geocoded households. Employing survival models with two alternative timescales and two control areas, the research measured how proximity to transit stations and project milestones affected relocation risks across different income levels. The analysis revealed a 4-5% increase in the exit probability for low- and middle-income households 5 to 9 years after the original line opened, whereas the newer extension saw a 6-10% decrease in outward migration for these groups during its first four years.

Why it matters — It demonstrates that transit-induced demographic shifts are driven more by selective in-migration of wealthier households than by immediate, disproportionate displacement of existing residents. Additionally, it establishes that displacement pressures can emerge with a significant lag of up to nine years after transit operations begin, indicating that mitigation policies must extend far beyond the initial construction phase.

Caveat: The findings are based on a single case study of two light-rail corridors in Charlotte, North Carolina, which may limit generalizability to cities with different housing markets or transit systems.

Transportation Research Part A Policy and Practice · doi

A total generalised cost assessment of e-scooter trips and their public transport competitiveness

This study calculated and compared the total generalised cost of shared e-scooter trips in Melbourne against their equivalent public transport alternatives. The analysis identifies specific spatial and temporal conditions where e-scooters outperform transit, particularly for non-radial journeys, trips near entertainment venues, and travel during late evening hours.

Why it matters — It maps the precise service gaps of public transport networks by demonstrating where and when micro-mobility options offer a more cost- and time-effective alternative for urban travelers.

Caveat: The findings are based on trip data from a single city, Melbourne, which may limit generalizability to cities with different transit layouts or pricing structures.

Transport Policy · pdf · doi

Everyday mobility practices of older women in Santiago, Chile: the importance of walking in the neighbourhood

This study analyzes the everyday mobility of 25 older women living in three central neighborhoods of Santiago, Chile, using semi-structured interviews and participatory mapping. The findings show that walking serves as the primary mode of transport for these residents, facilitating daily tasks, caregiving, and social connections with peers. Participants identified specific physical barriers in the built environment that restrict their movement, including narrow and uneven sidewalks, a lack of curb ramps, and insufficient public benches.

Why it matters — It provides empirical evidence from a Global South context on how neighborhood infrastructure directly shapes the independent mobility and social integration of older women, offering specific design targets to support aging in place.

Caveat: The findings are based on a small, qualitative sample of 25 women from a single city, which may not represent the experiences of older populations in other urban configurations.

Transportation Research Interdisciplinary Perspectives · pdf · doi

Navigating uncertainty: Safety solutions for autonomous vehicle integration into mixed-mode mobility

This perspective paper synthesizes research across neuroscience, robotics, and behavioral science to map how autonomous vehicles (AVs) and humans represent and act under uncertainty. It establishes a shared two-stage framework of control under partial information—probabilistic inference followed by risk-weighted action selection—to bridge the gap between mathematical risk models, robotic planning, and human psychological uncertainty.

Why it matters — It establishes human psychological uncertainty as a concrete, behaviorally relevant variable for AV engineering, providing a conceptual pathway to design self-driving systems that actively model and reduce the uncertainty of the humans sharing the road with them.

Transportation Research Interdisciplinary Perspectives · doi

Optimizing last-mile delivery in furniture logistics: A data-driven study on customer trust, perception, and retention

This study analyzed survey data from 400 respondents in Thailand using Structural Equation Modeling to evaluate what drives customer perception and behavioral intention for third-party last-mile furniture delivery. The model integrates SERVQUAL, trust-building, and customer engagement and retention frameworks to test their relative influence on customer choices.

Why it matters — The findings challenge the common assumption that service quality and trust are the primary drivers of customer satisfaction in premium logistics, demonstrating instead that these act as baseline expectations while relational engagement serves as the actual differentiator for customer retention.

Caveat: The study relies on self-reported survey data from a single country, which may limit how well these behavioral dynamics translate to non-premium logistics sectors or different cultural contexts.

Transportation Research Interdisciplinary Perspectives · pdf · doi

Islands in a sea of stone: Distinct microbial soil communities in Tilia tree pits compared to paved soils

Using DNA metabarcoding, this study characterized bacterial and fungal soil communities across five urban typologies, including open tree pits with Tilia trees, soils beneath sidewalks, and soils under roadways, both with and without nearby trees, across five Dutch cities. While bacterial alpha diversity remained consistent across all typologies, fungal richness was 91% higher near street trees and 304% higher in open tree pits compared to areas without trees. The presence of trees also correlated with an overrepresentation of putative fungal traits and a higher abundance of bacteria linked to nitrogen cycling.

Why it matters — The study demonstrates that street trees and their immediate pits act as critical hotspots for microbial biodiversity and functional redundancy, showing that even highly confined urban greenspaces can mitigate the ecological degradation caused by extensive soil paving.

Urban forestry & urban greening · doi

How vision-impaired riders navigate public transit disruptions: Challenges and opportunities

Through semi-structured interviews and thematic analysis with 12 blind and low-vision transit riders, this study documents how vision-impaired individuals navigate and recover from unexpected transit disruptions. The research identifies five key themes regarding how these riders anticipate predictable issues, adapt to unpredictable events, rely on transit operators and fellow passengers, and utilize specific recovery techniques.

Why it matters — It documents the specific, lived strategies and technological shortcomings experienced by vision-impaired riders during transit disruptions, providing empirical evidence to guide the design of more inclusive, real-time navigation technologies.

Caveat: The findings are based on a small qualitative sample of twelve riders within a single local transit system, which may not capture the full diversity of transit environments and infrastructure.

Journal of Urban Mobility · doi

Pedaling through space: Lefebvrian analysis of Turkish immigrants’ mobility experiences in the Netherlands

This study analyzes the mobility experiences of 21 Turkish immigrants living in the Netherlands using semi-structured in-depth interviews analyzed through thematic and inductive approaches. It frames their cycling practices through Lefebvre's production of space theory to examine how they build urban citizenship, adapt to a new mobility habitus, and navigate physical and social barriers like bicycle theft, infrastructure gaps, and normative pressures.

Why it matters — It demonstrates that cycling is not merely a utility or infrastructure choice for immigrants but a complex site of cultural negotiation, where adopting the practice serves as a marker of local belonging while simultaneously exposing systemic representational and epistemic inequities in cycling policy.

Caveat: The findings are based on a small, qualitative sample of 21 self-selected Turkish immigrants, which may not represent the diverse experiences of all immigrant groups or those living in different regions of the Netherlands.

Journal of Urban Mobility · doi

Examining the Associations between Visual and Non-Visual Elements and Cyclists' Route Choices for Various Trip Purposes

This study analyzes cycling trips in Montreal, Canada, to compare the actual routes chosen by cyclists against the shortest possible paths. It combines non-visual data, such as socioeconomic factors and two-dimensional spatial environments, with visual perception metrics derived from street-view images. The analysis reveals that cyclists deviate from the shortest path to prioritize routes with higher levels of greenery and lower levels of motorization.

Why it matters — It demonstrates how subjective visual experiences captured through street-level imagery, rather than just traditional two-dimensional network distance, actively shape route choices for different trip purposes.

Caveat: The findings are based on a single-city case study in Montreal, which may limit generalizability to cities with different cycling infrastructures or topographies.

arXiv · pdf

Joint Utilization of Geospatial and census proxies for Autoencoder-Assisted Downscaling (JUGAAD) of socioeconomic indicators in India

The researchers developed JuGAAD, a deep learning framework that downscales coarse socioeconomic indicators to fine-grained village-cluster scales. Using Indian census and National Sample Survey Office (NSSO) data from 2001 and 2011, the method averages geospatial and census data into intermediate tessellations, compresses high-dimensional survey data using an autoencoder, and maps the physical and demographic data to this latent representation to predict local poverty and food security indicators.

Why it matters — It provides a method to bridge the scale gap between high-resolution census data and coarse, infrequent socioeconomic surveys, allowing planners to estimate local poverty metrics at a fine administrative scale without waiting for new, expensive field surveys.

Caveat: The framework's predictions are validated against ground-truth indicators aggregated at the district level rather than direct village-level observations.

arXiv · pdf

From Feasibility to Desirability: Plan, Learn, Adapt (PLA) Framework for Personalized On-Device Itinerary Generation

The Plan, Learn, Adapt (PLA) framework generates personalized trip itineraries on-device by combining a heterogeneous ensemble of lightweight planners, a compact Bradley-Terry reward model trained on pairwise comparisons, and a local refinement stage. Evaluated on 2,519 human comparisons across more than 100 U.S. cities, the framework achieved a 67.8% win rate and 100% feasibility, whereas frontier LLMs like GPT-5, Claude Opus 4.5, and Gemini 3 Pro achieved 0% feasibility under the same constraints. Production deployment within the FlyEnJoy app demonstrated an average on-device latency of 109.9 ms and a 91% increase in itinerary completion rates.

Why it matters — It demonstrates that combining lightweight classical planners with compact reward models can guarantee 100% feasible, personalized itineraries on resource-constrained mobile devices, outperforming large language models which completely fail to meet hard combinatorial constraints.

arXiv · pdf

HiLLTS: Zero-Shot Hierarchical LLM-Guided Traffic Signal Control for Sustainable Transportation

The researchers developed HiLLTS, a hierarchical, three-layer Large Language Model framework for traffic signal control that coordinates vehicles across central, district, and cluster-level intersection agents. Tested under varying traffic demands, the zero-shot system reduced average commuter waiting times by 36.73% in low-congestion and 14.71% in high-congestion scenarios compared to the strongest non-LLM baselines, while simultaneously cutting CO2 emissions by 7.87% and 8.57% respectively.

Why it matters — It demonstrates that LLM-guided coordination can manage complex urban traffic networks in a zero-shot manner, bypassing the extensive retraining, simulation data, and complex reward engineering required by traditional reinforcement learning approaches.

Caveat: The performance gains and environmental metrics are evaluated within simulation environments rather than real-world physical deployments.

arXiv · pdf