Since its inception, the choice modelling field has been dominated by theory-driven approaches. Machine learning offers an alternative data-driven approach for behaviour and is increasingly drawing interest in our field. Cross-pollination of machine models, techniques practices could help overcome problems limitations encountered current paradigm, such as subjective labour-intensive search processes model selection, inability to work with text image data. However, despite potential benefits...

Public transport ridership around the world has been hit hard by COVID-19 pandemic. Travellers are likely to adapt their behaviour avoid risk of transmission and these changes may even be sustained after To evaluate travellers' in public networks during times assess how they will respond future pandemic, we conduct a stated choice experiment with train travellers Netherlands. We specifically related three criteria affecting transmission: (i) crowding, (ii) exposure duration, (iii) prevalent...

This study presents a new method to infer the average two-dimensional (2D) spacing between interacting vehicles in urban traffic from trajectory data. In this context, 2D reflects amount of road space consumed by pairs vehicles, and is related density at macroscopic level. Due complex interaction conflicts traffic, inherent assumptions traditional flow models, such as unidirectional homogeneity, are often violated. Such violation challenges direct measurement vehicle spacing. The proposed...

Artificial Neural Networks (ANNs) are rapidly gaining popularity in transportation research general and travel demand analysis particular. While ANNs typically outperform conventional methods terms of predictive performance, they suffer from limited explainability. That is, it is very difficult to assess whether or not particular predictions made by an ANN based on intuitively reasonable relationships embedded the model. As a result, for analysts gain trust ANNs. In this paper, we show that...

Informative representations enhance model performance and generalisability in downstream tasks. However, learning self-supervised for spatially characterised time series, like traffic interactions, poses challenges as it requires maintaining fine-grained similarity relations the latent space. In this study, we incorporate two structure-preserving regularisers contrastive of spatial series: one regulariser preserves topology similarities between instances, other graph geometry across temporal...

Background: Long-term noise annoyance can be expected to have worse outcomes than short-term annoyance. This study investigates over time, its association with personality traits and potential reciprocal effects between health Methods : Firstly, we conducted a Longitudinal Latent Class Analysis identify profiles. We further analysed the effect of Big Five on likelihood belonging these Secondly, used Cross-lagged Panel Models analyse whether changes in precede or vice versa. For both...

A thorough understanding of how urban space characteristics, such as equipment or network topology, affect people's density in spaces is essential to well-informed policy making. Hitherto, studies have primarily examined the characteristics impacts number people visiting different parts area (e.g., city center). However, these almost without exception used relatively small data sets, targeting specific neighborhoods places. As a result, their findings are confined areas and it unclear what...

We investigate the evolution of residential segregation patterns in Netherlands, with a focus on population non-western migration background. Unlike previous research relying predefined spatial structures, this study employs regionalization approach to track social enclaves 82 municipalities from 2015 2020. Enclaves have become more mixed historically homogeneous whereas other municipalities, they homogeneous. In addition, we find positive association between increase share background at...

Integrating Automated Vehicles (AVs) into existing traffic systems holds the promise of enhanced road safety, reduced congestion, and more sustainable travel. Effective integration AVs requires understanding interactions between Human-driving (HVs), especially during transition period in which HVs coexist a mixed environment. Numerous recent empirical studies find headways human drivers following an AV compared to HV, attribute this reduction behavioural changes when they follow AVs....

During the past 40 years, mobility patterns have enduringly changed several times as a result of occurrence number "substantial changes". Examples such substantial changes are rapid emergence affordable air travel, oil crises, and profound ICT developments. To most researchers policy-makers in transportation, it seems more than likely that next years will also witness changes, some which might even larger impacts on ones described above. However, literature their impact difficult to access...

Recent empirical evidence suggests that travellers are becoming increasingly multimodal. Coinciding with this trend, a growing interest can be observed in the transport literature to study concept of multimodality. Most studies, regard, have focused on assessing determinants multimodal travel behaviour. While it is interesting know which factors, at certain moment time, affect membership mono/multimodal patterns, one general omission current relates questions how and why switch between...

Fully Automated Vehicles (AVs) have been widely expected to revolutionise the future travel experience. Recent studies shown that their impact may also reach beyond episode, and lead users alter other activities performed during day – daily lifestyles. This study is among first empirically investigate changes travellers expect in with AVs. To this aim, we created an interactive stated activity-travel survey, which respondents designed current schedule and, following that, redesigned it while...