In the face of mounting pressures—including climate change—public authorities and researchers are actively exploring strategies to advance sustainable and active transportation modes. One critical challenge is the transition from car-dominated mobility to walkable urban environments, which promises to reduce congestion, lower pollution, and enhance public health. However, effective planning for this transition is hampered by the lack of reliable, city-wide data on pedestrian infrastructure—a data gap that impedes efforts to improve walkability and accessibility. To address this shortfall, we introduce a novel, deep learning-based scalable pipeline for extracting continuous, geocoded sidewalk networks from high-resolution aerial imagery. Our approach integrates advanced semantic segmentation with geometric and graph-based post-processing to convert raw imagery into structured and interconnected pedestrian networks. This work provides urban planners and policymakers with the robust data needed for informed decision-making and paves the way for more sustainable, human-centred urban mobility strategies.

Identifying factors that promote active mobility, especially walking, is essential for designing resilient and livable cities and promoting sustainable urban mobility. In spite of recent advances in this direction, available data often remains too spatially and temporally coarse, which constrains analysis. This paper leverages high resolution data from over 200 pedestrian count sensors, placed along Barcelona's commercial areas, providing a detailed understanding of how walking volume has evolved over the past five years, how it varies across neighborhoods, and which socioeconomic and urban attributes influence it. We find that while overall pedestrian traffic has increased, a neighborhood-scale analysis reveals a nuanced picture of fluctuations, including increases, declines, and periodic patterns. The use of global regression models allows us to identify seven key urban factors that shape pedestrian mobility. Subsequently moving the analysis to spatially-aware regression models, we identify the spatial non-stationarity of these factors across the city, indicating the presence of distinct behavioral groups within the urban population. The detailed spatial resolution of our findings provides municipal decision-makers with insights for implementing precise interventions and continually evaluating their effects. Moreover, monitoring pedestrian traffic before and after urban initiatives, while adjusting for seasonal, daily, and time-of-day variations, can yield critical insights for developing pedestrian-oriented urban environments.

In recent years, the design (and re-design) of cities to encourage walkability has taken on new urgency as part of a wider campaign for sustainable urban development. Complementary to other approaches like infrastructure improvements, increases in residential density, or traffic calming measures, here, we show how planning for walkability can be augmented by the adaptation of tools and approaches from the study of urban networks, by privileging the pedestrian perspective of short-distance access over the car (and rapid transit) perspective of flow and efficiency. Using a recently developed sidewalk network model that moves towards a more realistic representation of the pedestrian environment, we propose a framework for assessing multi-factor walkability using percolation theory and insights into pedestrian behavior. We apply our framework to the city of Barcelona, and show how it can be used to optimize service location and access for vulnerable populations (the elderly and young).

From a transport perspective, increasing active travel -and walking in particular- is crucial for the future of sustainable cities, as reflected in global decarbonisation policies and agendas. Further, walking is much more than a mere mode of transport: it provides a fundamental social function, fostering vibrant cohesive communities. Arguably, walking and its associated infrastructure -sidewalks- should rank among the highest priorities for planning authorities. However, efficiency- and speed-driven urbanisation has gradually reallocated street space to private cars, leading to automobiles being the prioritised mode of transport today. Empirical research has generally followed suit, and a systemic understanding of walking as a phenomenon is largely missing, i.e., questions like how connected, resilient, accessible, or socially equitable is the pedestrian infrastructure of whole neighbourhoods and cities. Such relative neglect of sidewalk network research is, first and foremost, the consequence of a generalised lack of publicly available data on sidewalk infrastructure worldwide. A second reason might be its apparent lack of interest from a systemic standpoint: pedestrian mobility does not produce coordination challenges on the scale that cars do. In this work, we confront this perception by showing that there is ample research potential in the study of system-wide sidewalk networks, with both structural and dynamical challenges which might be critical to pursue the latest aspirations towards sustainable mobility in cities.

Increased interaction between and among pedestrians and vehicles in the crowded urban environments of today gives rise to a negative side-effect: a growth in traffic accidents, with pedestrians being the most vulnerable elements. Recent work has shown that Convolutional Neural Networks are able to accurately predict accident rates exploiting Street View imagery along urban roads. The promising results point to the plausibility of aided design of safe urban landscapes, for both pedestrians and vehicles. In this paper, by considering historical accident data and Street View images, we detail how to automatically predict the impact (increase or decrease) of urban interventions on accident incidence. The results are positive, rendering an accuracies ranging from 60 to 80%. We additionally provide an interpretability analysis to unveil which specific categories of urban features impact accident rates positively or negatively. Considering the transportation network substrates (sidewalk and road networks) and their demand, we integrate these results to a complex network framework, to estimate the effective impact of urban change on the safety of pedestrians and vehicles. Results show that public authorities may leverage on machine learning tools to prioritize targeted interventions, since our analysis show that limited improvement is obtained with current tools. Further, our findings have a wider application range such as the design of safe urban routes for pedestrians or to the field of driver-assistance technologies.

Cities world-wide have taken the opportunity presented by the COVID-19 pandemic to improve and expand pedestrian infrastructure, providing residents with a sense of relief and pursuing long-standing goals to decrease automobile dependence and increase walkability. So far, due to a scarcity of data and methodological shortcomings, these efforts have lacked the system-level view of treating sidewalks as a network. Here, we leverage sidewalk data from ten cities in three continents, to first analyse the distribution of sidewalk and roadbed geometries, and find that cities present an unbalanced distribution of public space, favouring automobiles at the expense of pedestrians. Next, we connect these geometries to build a sidewalk network –adjacent, but irreducible to the road network. Finally, we compare a no-intervention scenario with a shared-effort heuristic, in relation to the performance of sidewalk infrastructures to guarantee physical distancing. The heuristic prevents the sidewalk connectivity breakdown, while preserving the road network’s functionality.

The overwhelming amounts of data we generate in our daily routine and in social networks has been crucial for the understanding of various social and economic factors. The use of this data represents a low-cost alternative source of information in parallel to census data and surveys. Here, we advocate for such an approach to assess and alleviate the segregation of Syrian refugees in Turkey. Using a large dataset of mobile phone records provided by Turkey’s largest mobile phone service operator, Türk Telekom, in the frame of the Data 4 Refugees project, we define, analyse and optimise inter-group integration as it relates to the communication patterns of two segregated populations: refugees living in Turkey and the local Turkish population. Our main hypothesis is that making these two communities more similar (in our case, in terms of behaviour) may increase the level of positive exposure between them, due to the well-known sociological principle of homophily. To achieve this, working from the records of call and SMS origins and destinations between and among both populations, we develop an extensible, statistically-solid, and reliable framework to measure the differences between the communication patterns of two groups. In order to show the applicability of our framework, we assess how house mixing strategies, in combination with public and private investment, may help to overcome segregation. We first identify the districts of the Istanbul province where refugees and local population communication patterns differ in order to then utilise our framework to improve the situation. Our results show potential in this regard, as we observe a significant reduction of segregation while limiting, in turn, the consequences in terms of rent increase.

In the wake of the pandemic, the inadequacy of urban sidewalks to comply with social distancing remains untackled in academy. Beyond isolated efforts (from sidewalk widenings to car-free Open Streets), there is a need for a large-scale and quantitative strategy for cities to handle the challenges that COVID-19 poses in the use of public space. The main obstacle is a generalized lack of publicly available data on sidewalk infrastructure worldwide, and thus city governments have not yet benefited from a complex systems approach of treating urban sidewalks as networks. Here, we leverage sidewalk geometries from ten cities in three continents, to first analyze sidewalk and roadbed geometries, and find that cities most often present an arrogant distribution of public space: imbalanced and unfair with respect to pedestrians. Then, we connect these geometries to build a sidewalk network --adjacent, but not assimilable to road networks, so fertile in urban science. In a no-intervention scenario, we apply percolation theory to examine whether the sidewalk infrastructure in cities can withstand the tight pandemic social distancing imposed on our streets. The resulting collapse of sidewalk networks, often at widths below three meters, calls for a cautious strategy, taking into account the interdependencies between a city's sidewalk and road networks, as any improvement for pedestrians comes at a cost for motor transport. With notable success, we propose a shared-effort heuristic that delays the sidewalk connectivity breakdown, while preserving the road network's functionality.

The Data for Refugees (D4R) Challenge resulted in many insights related to the movement patterns of the Syrian refugees within Turkey. In this chapter, we summarize some of the important findings, and suggest policy recommendations for the main areas of the challenge. These recommendations are sometimes broad suggestions, as the policy interventions involve many factors that are difficult to take into account. We give examples of such issues to help policy-makers.