Autonomous driving is an ongoing trend with the potential to revolutionize transport. In this paper, the authors investigated previous studies that focused on user preference and acceptance of autonomous vehicles. Some studies combined the choice of preferred traits and the intention of behavior to test the preferences of individuals. The research reviews focused primarily on behavioral intentions rather than technological preferences. Research documents on autonomous vehicles have been collected from various sources. The document's structure includes sections on methods, attributes, sociodemographics, attribute variables, identified gaps and conclusions. Discrete selection experiments are used to understand user preferences for new transportation technologies. The services of autonomous vehicles consider variables based on vehicles and individuals. Since 2013, the research on level 5 autonomous vehicles in Europe and the United States has focused on online surveys. Different models are used for data analysis to calculate individual heterogeneities in preferences. In some studies, hybrid selection and nested logit models were also used to capture user preferences. Experimental selection design in the literature often involves using orthogonal and D-efficient methods.

In railway project appraisal, the train traffic and complementing and competing modes’ supplies are one of the most important factors for estimation of travel demand and hence project benefits. Today, the forecast timetables that the Swedish Transport Administration (Trafikverket) uses for passenger forecasts are based on past developments, supplemented by an expert assessment of expected future supply developments. There are some doubts as to whether the method is sufficiently objective and consistent between different objects to provide a fair planning basis. The aim of this study is to develop a data-driven method for the supply that can be used as a starting point for travel forecast generation.

In this study, a model, Multi-Task Heterogeneous Graph Attention Neural Network (MT-HGATNN), was developed and applied to forecast train supply across multiple train categories and line sections, or station pairs. By leveraging structured timetable data, socioeconomic inputs, and scenario-based forecasts, the model provides accurate and interpretable insights into both current and future rail demand.

The model performs well in both retrospective validation (2023 data) and prospective forecasting (base year 2045), with model training on and a case study of East Coast line (Ostkustbanan). It is offering a framework for strategic transport planning under uncertainty. The multi-task learning approach enables joint modelling of multiple train categories, improving parameter efficiency and allowing the model to exploit latent interdependencies between tasks. The ability to integrate temporal granularity (hourly slots), spatial structure (station pair-level analysis), and diverse exogenous variables (e.g., GDP, car ownership, fares) further strengthens its applicability in complex real-world settings.

There are several promising directions for extending and deepening the work presented in this study. One important step would be to evaluate the generalisability of the developed MT-HGATNN model by applying it to other rail corridors or potentially at the national network level. This would help assess the model’s robustness across different geographical contexts and service supplies.

Vid utvärdering av järnvägsprojekt är tågtrafiken och kompletterande och konkurrerande färdmedelsutbud en av de viktigaste faktorerna för att förutse resefterfrågan och därmed projektnyttor. Idag baseras de prognostidtabeller som Trafikverket använder för resandeprognoser på hittillsvarande utveckling, kompletterat med en expertbedömning av förväntad framtida utbudsutveckling. Det finns vissa tvivel om huruvida metoden är tillräckligt objektiv och konsistent mellan olika objekt för att ge ett rättvisande planeringsunderlag. Syftet med denna studie är att utveckla en datadriven metod för utbudet som kan användas som utgångspunkt för generering av resandeprognoser.

I denna studie utvecklades och tillämpades en modell, Multi-Task Heterogeneous Graph Attention Neural Network (MT-HGATNN), för att prognostisera tågutbud över flera tågkategorier och linjedelar, eller stationspar. Genom att utnyttja strukturerad tidtabellsdata, socioekonomiska indata och scenariobaserade prognoser ger modellen korrekta och tolkningsbara insikter i både nuvarande och framtida järnvägsefterfrågan.

Modellen presterar väl i både retrospektiv validering (data från 2023) och prospektiv prognos (basår 2045), med modellinlärning på och en fallstudie av Ostkustbanan. Den erbjuder ett ramverk för strategisk transportplanering under osäkerhet. Metoden för fleruppgiftsinlärning möjliggör gemensam modellering av flera tågkategorier, vilket förbättrar parametereffektiviteten och gör det möjligt för modellen att utnyttja latenta beroenden mellan uppgifter. Möjligheten att integrera tidsmässig segmentering (timme för timme), rumslig struktur (analys på stationsparnivå) och olika exogena variabler (t.ex. BNP, bilägande, biljettpriser) stärker ytterligare dess tillämpbarhet i en komplex verklighet.

Det finns flera lovande riktningar för att utöka och fördjupa arbetet som presenteras i denna studie. Ett viktigt steg skulle vara att utvärdera generaliserbarheten hos den utvecklade MT-HGATNN-modellen genom att tillämpa den på andra järnvägskorridorer eller potentiellt på nationell nätverksnivå. Detta skulle bidra till att bedöma modellens robusthet över olika geografiska sammanhang och utbud.

Short-haul Flight (SHF) bans aim to stimulate the air-to-rail modal shift, consequently curbing the aviation sector’s environmental impact. We investigate the potential implications of various SHF ban policy designs on CO2-equivalent (CO2e) emissions, passengers’ travel times and rail capacity under the assumption of full air-to-rail modal substitution. Ranging from 0.4 Mt to 7.5 Mt CO2e, respectively 0.6% to 12.3% of the emissions of commercial intra-European aviation, the environmental impact of SHF ban policies is shown to be largely dependent on the policy design, namely the affected journey types and rail in-vehicle time thresholds. Our findings underscore the significant challenges of implementing such policies for the longer rail in-vehicle time thresholds and wider geographical scopes associated with noticeable environmental benefits. Despite the marginal impact of SHF ban policies on capacity utilisation in the case study, considerable interventions on rail infrastructure would be required to absorb existing air demand completely while ensuring attractive schedules. The results contribute to the ongoing policy debate, providing actionable insights to support Europe’s ambitious environmental goals in the transport sector.

To boost the market share of plug-in electric vehicles (PEVs) and ultimately achieve 100% of new vehicle sales, it is crucial for early adopters to continue choosing PEVs for their future vehicles. This underscores the importance of understanding who is more likely to be satisfied with PEVs and how this satisfaction impacts their future vehicle choices. This study investigates how satisfaction with the availability of public charging infrastructure, as well as with vehicle and charging technology, influences the future vehicle choices of Hungarian PEV users, utilizing a hybrid choice modeling (HCM) framework. Additionally, this study profiles the PEV users, taking their mobility and sociodemographic characteristics into account, and explores who are more likely to be satisfied with the availability of public charging infrastructure, as well as with vehicle and charging technology, using the structural equation models within the developed HCM framework. The results indicate that satisfaction with vehicle and charging technology as well as with the availability of public charging infrastructure has a positive impact on future PEV choice decisions. Furthermore, the findings reveal that PEV users who frequently charge at work and have a low average daily mileage are more likely to be satisfied with the vehicle and charging technology (e.g., charging time, range) than others. Commuters who use park-and-ride facilities and frequently charge their PEVs there are also more likely to be satisfied with the vehicle and charging technology. Those residing in Budapest who do not use their PEV daily are more likely to be satisfied with the availability of public charging infrastructure. Similarly, PEV users who are willing to walk at least 500 meters from their destination to reach a charging station are more likely to be satisfied with the availability of public charging infrastructure. Moreover, multi-car households and individuals over 40 are more inclined to choose a PEV for their next vehicle. However, household income and university education level did not significantly impact future PEV choice decisions. These insights can assist policymakers in profiling PEV users based on mobility and demographic factors, enabling the formulation of targeted strategies and investments to enhance satisfaction with PEVs and ultimately boost their adoption and market penetration.

Public transport service (PTS) analysis and provision is an important and challenging issue for public transport agencies. The results of the PTS analysis help transport planners to identify the areas in need of PTS improvement. Furthermore, relevant policy actions need to be determined for service provision to reach the desired level of PTS improvement in the identified areas. Without an appropriate decision support tool, planners need to apply several blind trials to find a policy action which improves the PTS in the examined areas. This paper introduces a data-driven decision support tool for PTS analysis and provision. The proposed framework combines a potentially large number of PTS measures while taking the correlation among the investigated measures into account and develops high-dimensional supervised classification models that predict the PTS levels for different policy actions. With this approach, planners can identify and prioritize the areas in need of PTS improvement, determine what policy actions should be targeted to improve the PTS in the identified areas, and predict the PTS impacts of these policy actions in the examined areas. The application of the proposed framework is demonstrated in detail through a case study of Budapest, Hungary, which is followed by a hypothetical policy implementation. The results show that mostly outskirts are in need of PTS improvement. Furthermore, the underlying reasons behind the areas with poor overall PTS are studied to target the relevant policy actions that improve the PTS in the identified areas. The PTS impacts of the targeted policy actions are studied by using the developed high-dimensional supervised classification models.

This paper investigates the intention to use private autonomous vehicles (PAVs) in developed (i.e., the United States, Belgium, the United Kingdom, Italy, Portugal, and Hungary) and developing countries (i.e., Egypt, Iraq, Jordan, Lebanon, and Saudi Arabia). Self-efficacy, the attitude toward using PAV technology, and the trust in PAV constructs are integrated into an extended Unified Theory of Acceptance and Use of Technology (UTAUT2) model to increase the understanding of users' perceptions of PAVs in developed and developing countries. The results of partial least squares structural equation modeling show the attitude toward using PAV technology to be the strongest predictor of the intention to use PAVs in developed and developing countries. Performance expectancy and trust in PAV are the strongest predictors of attitude toward using PAV technology in developed and developing countries, respectively. Effort-related constructs (i.e., effort expectancy and self-efficacy) do not directly affect the behavioral intention of the respondents from the developed countries while significantly affecting the behavioral intention of the respondents from the developing countries. The results of the mediation analysis show that the relationship between effort expectancy and behavioral intention is fully mediated by the attitude toward using PAV technology and performance expectancy for the respondents of the developed countries. Furthermore, the results of the multi-group moderation analysis show the relationship between performance expectancy and behavioral intention as well as social influence and the trust in PAV is not moderated by any of the moderators (i.e., age, gender, educational level, employment status, income level, car ownership status, and driving license status), neither for the developed countries nor for the developing countries.

The literature on autonomous vehicle (AV) acceptance highlights the significance of hedonic motivation in AV adoption. Numerous studies empirically confirm hedonic motivation as either the most or one of the most influential factors in the acceptance of AV. This fact calls for a need to achieve a wider understanding of the potential AV users’ perceived enjoyment (i.e., hedonic motivation). To this end, this study investigates the antecedents of hedonic motivation in the AV technology acceptance domain. The partial least square structural equations modeling approach was applied to analyze the data collected from 1823 respondents from 11 countries via an online survey. The developed hypotheses are examined for the entire sample, as well as separately for the Global North (GN) countries' sample, Global South (GS) countries' sample, and each individual country through a cross-country analysis. The results for the entire sample indicate that social influence is the strongest predictor of hedonic motivation, consistent with the findings of the GN sample. However, in the GS sample, self-efficacy emerges as the strongest predictor of hedonic motivation. Perceived safety is the second strongest predictor of hedonic motivation for both the GN and GS samples, highlighting its importance in relation to the perceived enjoyment of PAV. Trust does not significantly contribute to hedonic motivation, while the enjoyment of driving conventional cars has a small negative impact on hedonic motivation in the GS sample. The cross-country analysis reveals general patterns in the findings of the GN and GS samples, while highlighting a few exceptions. The results of the multi-group moderation analysis highlight the significant impact of the respondents' geographical distribution (GN vs GS) on their perceived enjoyment of PAV. Additionally, the analysis indicates that female respondents who enjoy driving conventional cars are less likely to perceive PAV as enjoyable compared to male participants who enjoy driving conventional cars.

Recently, in the literature, microscopic simulation is one of the most attractive methods in impact assessment of automated vehicles (AVs) on traffic flow. AVs can be divided into different categories, each having different driving characteristics. Hence, calibrating microscopic simulators for different AV categories could be challenging in AVs' impact assessment. The PTV Vissim microscopic traffic simulation software has been calibrated for simulating diverse types of AVs in a large body of literature. There are two main streams of studies in literature adapting AVs' driving behaviors in Vissim following either internal (i.e., adjusting the parameters of the Vissim's default driving behavior models) or external (i.e., adapting AVs' behavior through external VISSIM interfaces) modeling approaches. The current paper investigates how the PTV Vissim has been internally calibrated for the simulation of different types of AVs and compares the calibrated values in the literature with default values introduced in the recent version of PTV Vissim. In the present paper, the reviewed studies are partitioned into two main categories according to the characteristics of the studied AVs, the studies focused on autonomous automated vehicles (AAVs) and the ones focused on cooperative automated vehicles (CAVs). Our findings indicate that the literature expects a lower value for parameters including standstill distance (CC0), headway time (CC1), following variation (CC2), the threshold for entering "following" (CC3), negative/positive following thresholds (CC4/CC5), speed dependency of oscillation (CC6), oscillation acceleration (CC7), safety distance reduction factor (SDRF), and minimum headway front/rear (MinHW) for AVs than conventional vehicles (CVs). Besides, the literature expects higher values for parameters including standstill acceleration (CC8), acceleration at 80 km/h (CC9), looking distances, and maximum deceleration for cooperative braking (MaxDCB) for AVs. When cautious AVs are introduced, deterring effects are expected in the literature (e.g., higher CC0). Moreover, CAVs can have higher looking distance values compared with AAVs.