Understanding multi-modal urban mobility patterns is essential for effective planning and policy-making. Traditional data sources, such as infrequent surveys or smart card records, often lack the temporal, spatial, and modal comprehensiveness required to fully capture the complexity of multi-modal travel behavior. Emerging mobility data sources are instrumental in capturing these patterns and in enabling additional insights. This study leverages a digitally collected trajectory-level dataset (i.e., TravelSense) obtained from a smartphone application operated by the public transport authority of Helsinki, Finland. Unlike conventional public transport data, TravelSense provides insights into modal choices alongside temporal and spatial travel characteristics. In order to analyze mobility patterns and explore the capabilities of this novel dateset, a Latent Profile Analysis is employed to classify travelers based on these attributes over a week-long period, with profiles compared across three consecutive years (2022, 2023, and 2024). Findings reveal that while spatial travel patterns remain relatively stable, temporal and modal patterns exhibit greater variability. A distinct shift is observed between 2022 and subsequent years, likely reflecting post-pandemic behavioral changes. Key traveler groups identified include exclusive active mode users (13 % annually) and non-private car users, whose share declined from 38 % in 2022 to approximately 20 % in 2023 and 2024. Study findings offer valuable input for shaping evidence-based mobility policies, particularly those aiming to support sustainable travel behavior and adapt to evolving urban mobility needs through enhanced multi-modality. TravelSense enables detailed analysis of temporal, spatial, and modal travel patterns, underscoring the value of novel data for multi-modal transport research.

Providing relevant information is crucial in public transport systems. With the rise of digital passenger information systems (PIS), personalization has emerged as a means to meet passengers’ information needs better. To better understand how personalization can be implemented in PIS, five levels of personalization have been identified in the literature, highlighting varying degrees of system autonomy and passenger involvement. While these levels have advanced the understanding of personalization, their practical application remains limited. This paper builds upon an existing framework of personalization levels. It introduces an evaluation tool composed of distinct performance measures to help PIS developers assess their system’s current personalization level and identify opportunities for technological advancement. The tool enables a comparison of the personalization functionalities against the best practices defined by the personalization levels. The paper further outlines the creation of the tool through functional benchmarking, introduces system behaviors across levels, and evaluates commercial PIS through case studies, offering actionable insights for advancing PIS personalization.

This research introduces a new method for constructing and training an Urban Space Embedding Model (USEM) by integrating human perceptions and street-level images (SLI) into its formulation. Traditional urban embedding models often overlook subjective human experiences, such as perceptions of safety or attractiveness. To address this gap, our method leverages similarity judgments from over 1500 participants, who compared different urban spaces based on SLI. These human judgments were then used as a supervision signal in training the USEM, allowing the model to capture both visual and perceptual information about urban spaces. The method is implemented across the Netherlands, using around one million geo-tagged SLI, and demonstrated in Rotterdam. This approach represents a significant advancement in urban computing by incorporating human-centered data into urban modeling. It offers new opportunities for city planners and policymakers to better understand how urban spaces are perceived and to consider these perceptions in efforts to design more livable and inclusive environments.

Ride-hailing has become an important part of the urban mobility landscape. The main contribution of this study is to estimate how travellers perceive time when using ride-hailing compared to using conventional public transport, to better understand ride-hailing mode choice. We combine two unique datasets containing actual, individual passenger behaviour for the Washington DC area from October 2018: a large set of almost 250,000 individual ride-hailing trips made using Uber, and more than 326,000 public transport trips obtained from automated ticketing data. Contrary to previous studies our model estimations rely on over half a million directly observed passenger choices between ride-hailing and public transport, based on which we estimate a discrete choice model to infer travel time perceptions for both modes using a binomial logit model. Our results show that on average ride-hailing in-vehicle time is perceived 35% less negative than public transport in-vehicle time. We also found that waiting time for ride-hailing is valued 1.3 times more negative than ride-hailing in-vehicle time, which is about 20% less negative than the ratio between waiting and in-vehicle time found for public transport. Our study enables a more accurate modelling of ride-hailing by using mode-specific travel time coefficients derived from large-scale empirical data, which can improve the accuracy of modelling outputs and thus improve decision-making processes.

This dataset contains the outputs of the paper "Replacing short-haul flights with train travel: Exploring impacts, capacity requirements and policy implications", published in Transport Policy. The paper investigates the potential implications of various SHF ban policy designs on CO2-equivalent emissions, passengers’ travel times and rail capacity, under the assumption of full air-to-rail modal substitution. The datasets are published to substantiate paper results and for use by other researchers. The data made available in this repository refers to travel time and emissions assessments, including some inputs and the aggregated outputs. Due to the limitations agreed upon with the data provider, the input data is not complete. The input data has been collected from multiple sources, including OAG, Rome2Rio and Google Distance Matrix API, and has been cleaned and processed using Python. More information is available in the README.txt file and the paper referenced below.

As a two-sided digital platform, ride-sourcing has disruptively penetrated the mobility market. Ride-sourcing companies provide door-to-door transport services by connecting passengers with independent service suppliers labelled as “driver-partners”. Once a passenger submits a ride request, the platform attempts to match the request with a nearby available driver. Drivers have the freedom to accept or decline ride requests. The consequences of this decision, which is made at the operation level, have remained largely unknown in the literature. Using agent-based simulation modelling on the realistic case study of the city of Amsterdam, the Netherlands, we study the impacts of drivers’ ride acceptance behaviour, estimated from unique empirical data, on the ride-sourcing system where the platform applies regular and surge pricing strategies, and riders may revoke their requests and reject the received offers. Furthermore, we delve into the implications of various supply–demand intensities, a centralised fleet (i.e., mandatory acceptance on each ride request) versus a decentralised fleet (i.e., ride acceptance decision by each driver), ride acceptance rates, and surge pricing settings. We find that the ride acceptance decision of ride-sourcing drivers has far-reaching consequences for system performance in terms of passengers’ waiting time, driver's revenue, operating costs, and profit, all of which are highly dependent on the ratio between demand and supply. As the system undergoes a transition from undersupplied (i.e., real-time demand locally exceeds available drivers) to balanced and then oversupplied state (i.e., more available drivers than real-time demand), ride acceptance decisions result in higher income inequality. A high acceptance rate among drivers may lead to more rides, but it does not necessarily increase their profit. Surge pricing is found to be asymmetrically in favour of all the parties despite adverse effects on the demand side due to higher trip fare. This study offers insights into both the aggregated and disaggregated levels of ride-sourcing system operations and outlines a series of transport policy and practice implications in cities that offer such ride-sourcing systems.

We investigate the correspondence between network-based public transport network (PTN) supply indicators and passenger demand at the node level, by systematically assessing correlations between node centrality measures and passenger boarding counts across different graph representations of PTNs. At the stop-level, undirected L- and P-space representations with three different edge weightings: unweighted, service-frequency-weighted, and in-vehicle-time-weighted are analysed. In each case, we calculate degree, closeness, betweenness and eigenvector centralities and examine the relation shapes. At the route level, we examine degree and eigenvector centrality for unweighted and weighted C-space representations. We introduce a modified C-space representation with self-loops, with service frequencies as self-loop weights, and propose eigenvector centrality as a route-level supply indicator. Stop- and route-level properties are integrated using the B-space representation. This methodology was applied to a case study for a bus PTN in Ljubljana, Slovenia. Results show strong correspondence between passenger demand and degree and eigenvector centrality scores in the frequency-weighted P-space (correlation ≈0.7−0.8). Notably, the relationship between eigenvector centrality and passenger counts in the new C-space representation with self-loops exhibits logarithmic behaviour. Furthermore, the results suggest a minimum eigenvector centrality threshold (≈10⁻³) for a route to start facilitating passenger use. The route-level results from the B-space analysis show exponential convergence of passenger counts to route eigenvector centrality. Results of the stop-level analysis are in line with previous research and deepen the understanding of centrality measures as supply indicators. Most significantly, the route-level analysis is novel, and the results open promising venues for further research.

We present a method to classify street networks using only geo-tagged street-level imagery. By combining pre-trained image embeddings with unsupervised clustering, it produces visually coherent street typologies without supervised training or labeled data and requires only minimal data curation. The approach is lightweight, scalable, and, in principle, transferable across urban contexts. In a Delft (Netherlands) case study, we classify approximately 2,000 road sections using over 70,000 images. Our method recovers distinct street types such as residential, arterial, and historic ones. These results show that pre-trained visual embeddings alone can support effective street classification from visual inputs, offering a practical tool for urban planning, transport analysis, and mobility research.

Modular vehicles (MVs), equipped with autonomous driving, communication, and platooning capabilities, are emerging as a promising innovation in transportation, offering the potential to enhance operational efficiency, flexibility, and environmental sustainability. However, challenges and barriers to their successful implementation are not yet fully understood, which limits the realization of these benefits. This literature review synthesizes existing research on MVs across various applications, including passenger and freight transport, to provide a systematic evaluation of state-of-art, opportunities and challenges for modular freight transport systems. The review identifies research gaps in five areas, such as their integration with multimodal transportation, and highlights key deployment challenges including regulatory hurdles, human factors, financial constraints, and operational complexities. Our findings emphasize the need for policy development, system design research and further empirical validation to assess the practical feasibility and impacts of MVs in the freight transport sector.

The emergence of social media offers unprecedented opportunities to map social unrest with high spatiotemporal resolution. This study leverages geolocated social media footage to analyze the spatiotemporal distribution of the 2023 ‘Nahel Merzouk’ riots in France. Using a fine-tuned computer vision model, we detect riot-related content in visual data and validate our approach by comparing the spatiotemporal patterns of detected posts with rioting events reported in the press. Our method yields a spatial resolution of 300 300 m, thereby facilitating a detailed analysis of riot distributions at unprecedented scale. By applying density-based clustering, we map riots across seven French cities, revealing their highly localized and bursty dynamics. This study opens pathways for future research on the causes and dynamics of social unrest, enabling a deeper understanding of urban riots and their potential mitigation.

Bert van Wee, professor in Transport Policy at Delft University of Technology, the Netherlands, faculty Technology, Policy and Management, is retiring. Given his large contributions to EJTIR as Editor-in-Chief, editorial board member, author and reviewer, this Editorial Note is dedicated to his work for EJTIR. Over 25 years, van Wee published 18 papers and 1 book review in EJTIR covering a wide range of topics from road pricing to urban rail transport, vehicle automation and port throughput. What his studies have in common is that they explore how transport policies affect land-use and travel behaviour, as well as the economic and wider societal impacts of those policies. Bert van Wee’s generalist view on the transport system is rare, but, given the rising complexity of the system, increasingly needed to indeed be able to address the future challenges.

This study proposes three mathematical programming models with distinct optimization objectives for transfer optimization in a bi-modal public transport network. To improve the applicability of the models and expedite the solution process, some acceleration techniques, including eliminating redundant constraints and incorporating valid inequalities, are suggested. The models and solution methods are applied to a small toy network and a real-life bi-modal public transport network. The results indicate that compared to the third model, the second model can reduce the total transfer waiting time by 12.29% to 30.31%, while the longest transfer waiting time may increase by 4.35% to 22.22%. Furthermore, the third model, which prioritizes minimizing the longest transfer waiting time, may increase the total or average transfer waiting time. The results suggest that decision-makers need to make a trade-off between reducing total passenger transfer waiting time (for efficiency) and reducing the longest passenger transfer waiting time (for fairness). .

The electrification of transportation has emerged as a key focus area over the past decade, driven by the rise of electric vehicles (EVs) and supportive governmental policies. Conventional EV charging solutions, while foundational, face notable challenges such as high infrastructure costs, low flexibility, and underutilization. Simultaneously, emerging transportation modes such as autonomous vehicles, shared mobility, modular systems, and aerial vehicles, introduce additional complexities, demanding more innovative charging solutions. This review emphasizes the potential of charge-on-the-move systems referred to as dynamic charging, as a transformative approach to address these challenges. Dynamic charging enables EVs to recharge while in motion, presenting opportunities to minimize battery sizes, reduce emissions, and optimize operational efficiency. The study critically evaluates state-of-the-art dynamic charging technologies, including their benefits, limitations, and applicability to future mobility systems, while also comparing these solutions based on infrastructure costs, readiness, and scalability. The findings suggest that the future of EV charging will likely involve a hybrid approach, integrating both conventional and dynamic solutions. Key priorities for advancing dynamic charging include developing optimization models for infrastructure deployment, finding the balance between battery size and battery life, establishing interoperability standards, and enhancing energy transfer efficiency while ensuring safety and sustainability. By addressing these research challenges, dynamic charging systems have the potential to redefine EV infrastructure and support the broader transition to sustainable and efficient mobility ecosystems. This review serves as a guide for researchers and planners seeking to align charging technologies with evolving transportation needs.

Demand prediction is essential for effective management of Mobility-on-Demand (MoD) systems, as accurate forecasts enable better resource allocation, reduced wait times, and improved user satisfaction. Beyond that, probabilistic prediction methods that explicitly account for uncertainty are particularly valuable, as it allows decision-makers to assess risk and make robust plans under uncertain operational environments. However, most existing approaches focus on point predictions, which fail to capture the full spectrum of possible future outcomes. For probabilistic prediction, many methods typically rely on strong parametric distributional assumptions that may not accurately reflect the complex real-world environments. Nonparametric methods proposed in the literature, although promising, often suffer from high computational costs and model complexity, limiting their practical applicability. To overcome these challenges, we propose the Spatial-Temporal Graph Convolutional Network Variational Autoencoder (STGCN-VAE), a novel deep learning framework designed for uncertainty-aware probabilistic travel demand prediction in MoD services. The STGCN-VAE effectively captures complex spatial-temporal dependencies and inherent uncertainties in MoD demand data, generating diverse and realistic future demand scenarios and constructing comprehensive demand distributions. Specifically, the proposed framework integrates three key components: a Spatial-Temporal Graph Convolutional Network (STGCN) to learn complex spatial-temporal dependencies, a Variational Autoencoder (VAE) to compress these patterns into a latent space, and a Kernel Density Estimation (KDE) module to accurately construct probabilistic demand distributions and quantify uncertainties. Experiments on four different real-world MoD datasets including both rideshare and bikeshare services across different cities demonstrate that STGCN-VAE consistently outperforms state-of-the-art baselines in both point and probabilistic prediction, highlighting its robustness and broad transferability across service modes and urban contexts.

Today, there is growing concern over the climate impact of long-distance travel – academic conference travel included. While institutions worldwide develop sustainability policies, it remains unclear whether academics are willing to adopt such measures and change their behaviours for environmental reasons. Based on a survey of 245 Spanish academics, we identify a set of motivational factors underlying individuals’ intentions towards conference travel. Using Latent-Class-Cluster-Analysis, we detect three distinct profiles: “Low-concerned academics”, with limited regard for environmental concerns and social norms; “Performance-oriented academics”, driven by career advancement; and “Flight-shamed academics”, highly aware of environmental and social issues. Additionally, we examine perceptions of the actual, minimum, and expected number of conferences attended annually, recognising notable discrepancies. Our findings can inform targeted policies in four key areas: promoting greener choices, reshaping academic culture to normalise lower-carbon practices, raising awareness and strengthening institutional engagement, and supporting online participation.

Widespread congestion in metro systems often hinders passengers from boarding the first arriving train, making them compelled to adopt an alternative route, some of which involve travelling backwards. While this travel strategy has direct consequences for forecasting passenger flow distribution in congested networks, little is known about the travelling backwards phenomenon and why people adopt this travelling behaviour. The aim of this study is to understand passengers’ perception of time in various segments considering travelling backwards. To achieve this, we develop a route choice model using revealed preference data from smart card records. We find that passengers exhibit a greater aversion to waiting time and onboard time while travelling backwards. Specifically, passengers perceive each minute spent waiting on the turn-back stations’ platform as equivalent to 1.97 min on the origin platform. Similarly, each minute spent onboard the backwards train is perceived as equivalent to 1.24 min on the forwards train. Ignoring this difference in perception would result in the underestimation of the expected social benefits of demand management policies. Finally, we assess the potential benefits of travelling backwards under various passenger flow conditions, offering valuable policy insights regarding whether and how this behaviour should be regulated or promoted.

Metro networks face operational challenges due to increasing ridership and system growth, particularly in managing delay propagation. Epidemiology models have recently been an interesting method in transportation research for studying delays. This study, therefore, aims to investigate if the Susceptible-infectious-susceptible (SIS) model is suitable to help model delay propagation in a metro network through its ability to reproduce the vulnerability of metro stations for specific instances. Using data from the Washington Metro Network, two groups of delay propagation instances were selected and used for model training and testing using a differential evolution algorithm. The results indicate that the vulnerability values as calculated from the reallife data do not follow the expected trend. Still, our model can capture this variation with good vulnerability estimation accuracy for both groups. Also, the predicted vulnerability values for the first group are more accurate than for the second group. However, limitations such as underestimation and overestimation of station vulnerabilities, and sensitivity to training data were observed. These challenges stemmed from the dynamics between specific parameters and the lack of additional factors.

Recent advances in automation have accelerated the development of autonomous electric vehicles (AEVs), which offer the potential for continuous operation, constrained primarily by the need for recharging. We propose a dynamic charging strategy based on Mobile Autonomous Charging Pods (MAPs), which are battery-equipped electric vehicles capable of transferring energy to AEVs while in motion. We introduce a dedicated simulation framework within the microscopic traffic simulator SUMO, incorporating MAP-specific modules for assignment, navigation, and real-time energy transfer under realistic traffic constraints. We model the behavior of both MAPs and AEVs in a stylized looped network and evaluate system-level performance under various demand and fleet configurations. Key performance indicators include energy consumption, charging efficiency, battery utilization, and reductions in AEV battery capacity requirements. Simulation results demonstrate that MAPs can effectively support continuous AEV operation, achieving up to 14% battery downsizing with minimal infrastructure investment, while also reducing travel time by 7%, relative to fixed charging solutions. This study lays the foundation for simulation-based evaluation of MAP-based dynamic charging as a scalable, flexible, and efficient alternative to fixed charging solutions.

Railway systems suffer from disturbances in operations, such as extended section running times caused by temporary speed restrictions and prolonged dwell times at stations due to unexpected passenger volumes. These disturbances cause deviations from the original timetable and negatively impact service reliability and passenger experience. Effective and timely rescheduling measures are crucial in reducing the impact of these disturbances. Existing timetable rescheduling models that rely on optimization-based methods often struggle with computational inefficiency, especially when dealing with scenarios involving a large number of train services. To address these challenges, we propose a learning-based timetable rescheduling framework that considers scalability in its formulation to reduce the growing computational burden associated with an increasing number of train services. The proposed framework decomposes the complex rescheduling problem into multiple subtasks, facilitating a systematic approach to managing extensive railway networks with numerous stations and train services. A high-level agent, functioning as a centralized traffic controller, is responsible for decomposing the overall deviation reduction task into subtasks at a low level and assigning them to individual train services with the primary objective of minimizing the time required to restore the original timetable. Low-level agents, acting as distributed train dispatchers, are tasked with rescheduling the timetables of their assigned trains. These low-level agents employ various dispatching strategies, facilitated by inter-train communication, to search for optimal rescheduling solutions while adhering to operational constraints such as minimum headway requirements. The lowlevel agents utilize an actor-critic architecture to generate continuous control decisions for dwell and running times, enabling them to learn and optimize their performance. Knowledge-sharing mechanisms amongst the low-level agents enable faster and more robust learning. Furthermore, advanced exploration methods are integrated to enhance the efficiency of the agents' training process.

As concerns about climate change increase, the environmental impact of long-distance travel – including academic conference travel – is coming into focus. Multiple universities have recently started to deploy sustainability policies committed to net-zero targets. However, it remains uncertain today whether academics are ready to embrace such initiatives and transform their practices for the sake of the environment. In this study, we explore the motivational factors behind academics’ willingness to limit their conference travel based on a survey conducted in Spain. The results highlight the role played by a set of demographic, socioeconomic, work-related, and attitudinal factors, as estimated by means of an ordered logit model. All else being equal, postdoctoral researchers and individuals who live in single-person households are more likely than others to reduce their travel. In terms of psychological attributes, we detect that individuals with a higher level of green values and more influenced by social norms are more intended to limit their conference trips. Conversely, those who believe that conferences are a driver for professional development are less willing to lower their travel. Our findings can help institutions to identify the segments of academics with a higher (and lower) probability of changing their behaviour towards more sustainable habits.