Working from home (WFH) has changed commuting behaviour in the Netherlands, with office attendance increasingly concentrated on Tuesdays and Thursdays. This concentration poses a key challenge for Dutch peak-spreading policy, yet existing research consistently measures working from home as a frequency rather than a weekly structure, leaving the day-specific organisation of commuting behaviour largely unexplored. This study addresses this gap by conceptualising working from home as a weekly structure, applying latent class analysis (LCA) and latent transition analysis (LTA) to longitudinal survey data from the Landelijk Reizigersonderzoek (LRO). The data cover three annual waves (2023-2025) with a balanced panel of 1,026 respondents. Three distinct commuting profiles were identified: the Moderate Commuter (MC, 57%), the Intensive Full-Week Commuter (IFW, 30%), and the Tuesday and Thursday Commuter (TT, 13%). These profiles remained highly stable over time, with the vast majority of individuals staying in the same profile across consecutive measurement waves. None of the nine contextual policy factors examined reached statistical significance. However, directional patterns suggest that perceived improvements in working from home possibilities and public transport frequency are most strongly associated with transitions away from peak-day commuting among TT commuters. At the same time, perceived improvements in these factors reinforce peak-day concentration among IFW commuters. These findings suggest that generic improvements to working conditions or infrastructure are insufficient to redistribute peak commuting demand. Effective peak spreading may require targeted, day-specific interventions directed at the groups that contribute most to Tuesday and Thursday concentration. ... Read more

Temporary roadworks increasingly disrupt commuting in the Netherlands, yet behavioural adaptation is still weakly represented in roadworks appraisal. This paper examines how disruption severity reshapes commuter responses, with particular attention to continued car use, public transport (PT) switching, and working from home (WFH). The analysis is based on an online survey among working commuters and combines an attribute-based stated-choice experiment, a commute-timepivoted stated-choice experiment, and a revealed-preference module on experienced major disruption (N = 180). The results show that adaptation to roadworks severity is non-linear: modest increases in car travel time are often tolerated, whereas stronger disruption triggers substantially more substitution away from routine car commuting. WFH emerges as the dominant substitute when feasible, while PT plays a secondary but meaningful role. Work-related constraints, especially WFH feasibility and on-site obligations, strongly condition the adaptation channel. Revealed responses indicate stronger car persistence than stated scenarios suggest, pointing to an intention–behaviour gap. The findings imply that roadworks demand reduction should not be treated as a uniform residual effect, but as the outcome of multiple behavioural response channels. Index Terms—temporary roadworks, disruption severity, commuter adaptation, mode choice, working from home, behavioural heterogeneity, stated-choice experiment ... Read more

Peak-hour congestion remains a persistent challenge in many countries, including the Netherlands. A key factor behind congestion is the timing of daily commuting trips. Although many studies have examined why commuters choose specific departure times, most analyses focus on average behaviour. In reality, commuters often vary their departure times from day to day, and these variations differ strongly between individuals. Understanding this variability is important for designing policies that aim to shift travel away from peak hours.

This thesis investigates departure-time variability among Dutch car commuters using high-resolution travel data from the Nederlands Verplaatsingspanel (NVP). The dataset contains GPS-based travel records that make it possible to observe actual commuting behaviour across multiple days. To capture both daily patterns and differences between individuals, a multi-level Latent Class Cluster Model (ML-LCCM) is applied. This modelling approach groups commuting days into clusters with similar timing patterns, while simultaneously classifying individuals into broader behavioural profiles based on the variability in their daily schedules.

The results reveal that departure-time behaviour is highly heterogeneous. Several distinct day-level patterns emerge, ranging from highly regular commuting days to days with large timing variability. At the person level, different types of commuters can be identified, including individuals with very stable commuting schedules as well as those with much more flexible and irregular travel patterns. These behavioural profiles suggest that commuters differ substantially in their ability or willingness to adjust departure times.

The findings highlight that policies aimed at reducing peak-hour congestion should account for these differences in flexibility between commuters. Measures that assume all travellers can easily shift their travel times may therefore have limited effectiveness. By identifying distinct behavioural profiles, this research contributes to a more nuanced understanding of commuter behaviour and provides insights that can support the design of more targeted mobility policies. ... Read more

In tight labour markets, recruitment difficulties are often attributed to skill shortages, overlooking the critical role of physical location. Even if qualified candidates exist, vacancies may remain unfilled if the workplace is inaccessible. This issue is particularly acute in economic clusters like the Port of Rotterdam, where 88% of employers cite accessibility as a recruitment barrier. This suggests that labour shortages reflect not only a lack of talent but also a spatial and modal mismatch between job locations and the workforce.

Shifting the analytical lens from the traditional residential perspective to the employer perspective, this thesis investigates how public transport accessibility functions as a selective mechanism for commuter profiles and explains regional variations in vacancy fulfilment. The study combines a national Latent Class Cluster Analysis (LCCA) to identify socio-demographic commuter profiles with a spatial stepwise regression in the Rijnmond region to model vacancy duration and vacancy rates.

The results demonstrate that workplace accessibility fundamentally shapes the intake of commuters. High-accessibility locations attract highly educated, transit-oriented workers, whereas peripheral industrial areas draw car-dependent profiles. This creates a modal mismatch for urban workers without car access, a pattern observed nationally but intensified in the port.

Furthermore, the regression analysis reveals a distinct spatial paradox regarding recruitment efficiency. Public transport accessibility is a significant predictor for vacancy duration, explaining 18.2% of the spatial variation when interacting with urban density. In peripheral and industrial zones, better accessibility significantly reduces vacancy duration, acting as a critical solution to friction. Conversely, in dense urban cores, high accessibility is associated with longer vacancy durations, likely due to increased employer selectivity in larger labour pools. In contrast, accessibility showed minimal explanatory power for vacancy rates, which appear driven by structural, sector-specific shortages rather than connectivity.

This thesis concludes that accessibility acts as a selective filter, effectively excluding the urban talent vital for the energy transition from industrial employment. To bridge this gap, policy should prioritize smart, demand-responsive mobility and the strategic relocation of non-place-bound functions to urban hubs, rather than solely expanding traditional infrastructure. ... Read more

Distracted driving is a significant road safety issue, with roadside advertising known to compete for drivers’ attention and potentially affect driving performance. To mitigate to potential influence on crash risk by roadside advertising, policy guidelines are in place. This thesis aims to identify the current trends in evidence regarding driver distraction and roadside advertising for ultimately revising and strengthening the Dutch framework for the placement of roadside advertising along highways. This is done by updating the scientific evidence base, reviewing international best practices, and assessing the safety impact of digital billboards. A literature review identified both new and confirmatory insights that can strengthen the evidence‑base for Dutch guidelines. Policy documents from comparable well-performing road-safety countries were also analyzed, revealing that while most frameworks are grounded in scientific evidence, their implementation varies due to pragmatic considerations. To explore the extent of potential safety effects, a before-and-after analysis was conducted on 20 Dutch digital billboards using crash data from 2014 to 2024. Results were mixed at individual sites but showed a statistically significant positive safety effect at an aggregated level, though cautious interpretation is needed, due to several methodological limitations. The thesis concludes that despite the lack conclusive evidence on risk figures, well-studied safety performance indicators and international guidance can be used for revising the Dutch framework on advertising objects along highways. ... Read more

Human judgments are subject to a wide range of cognitive biases. Although decision-making methods are designed to support and structure the decision process, they rely on human judgments as critical inputs, and decision outcomes may therefore still be biased. This dissertation investigates how four cognitive biases, anchoring bias, framing effect, loss aversion, and status quo bias, affect different stages of multi-attribute value theory and the resulting outcomes, and how these influences can be mitigated. The findings reveal a fundamental duality of multi-attribute decision-making methods: while their structured procedures can introduce or reinforce cognitive biases, they also create opportunities to mitigate them. ... Read more

Travel behaviour is in a state of transition, towards a sustainable transport system. This dissertation examines how people make travel-related decisions, using longitudinal data and statistical models. The substantive focus is on working from home, travel-related attitudes, and external influences such as the weather and life-events. The results improve our understanding of decisionmaking within travel behaviour, enabling policy makers to make betterinformed decisions in the face of the transition. ... Read more

Organizations in complex and interdependent sectors, such as the aviation industry, are increasingly dependent on innovation networks to address major transitions. Although network governance and orchestration are key in steering such networks, research rarely examines both in innovation contexts. This qualitative case study explores how Schiphol Airport and its external innovation partner coordinate and manage collaboration in Schiphol’s innovation network. The findings show that a hybrid governance structure, combined with adaptive orchestration practices, ensures alignment, trust, and facilitates collaboration in the innovation network. The study shows how different collaboration settings require different orchestration approaches and offers practical steps to strengthen collaboration. ... Read more

Public transport reliability is often undermined by two interrelated factors: in-vehicle crowding and headway variability. While both phenomena are widely recognized, limited empirical research has quantified their combined effect in high-frequency surface transport systems, particularly under conditions of overlapping routes. This study analyzes over 1.7 million stop-level observations from the Geneva bus and tram network, integrating Automatic Passenger Count (APC), Automatic Vehicle Location (AVL), and GTFS schedule data. A random-effects ordered logistic regression model is applied to estimate how operational irregularities and network design features shape crowding levels, measured through a five-level index. Results show that headway variability is the strongest predictor of in-vehicle crowding: vehicles arriving late absorb accumulated demand and are significantly more likely to exceed design capacity. Route overlap amplifies this effect by introducing interdependencies between lines, leading to higher crowding levels in shared segments. Vehicle type also plays a role, with trams more frequently operating under crowded conditions, while trolleybuses remain less affected. These findings provide actionable insights for operators: interventions such as holding control, coordinated scheduling across overlapping lines, and targeted stop-level monitoring can mitigate the most severe crowding events without major infrastructure investments. The study contributes to transport reliability research by providing large-scale empirical evidence of the crowding–headway–overlap nexus, and offers practical guidance for agencies aiming to improve comfort and service quality in urban surface transport systems. ... Read more

As corporations face increasing pressure to meet sustainability targets and alleviate regional traffic congestion, understanding employee commuting behavior has become a critical management concern. Traditional mobility policies often fall short due to the diverse and complex motivations behind commuting choices. This thesis investigates how different employee segments in the Netherlands perceive enablers and barriers in their commuting decisions, aiming to inform more targeted and effective mobility strategies.

A mixed-method approach was employed. Quantitatively, the study analyzed the Landelijk Reizigersonderzoek (LRO) dataset using Latent Class Analysis (LCA), which is well-suited for identifying commuter segments based on categorical survey data. This analysis revealed distinct profiles among car, public transport, and bicycle users, highlighting significant variation even within the same mode of transport. For instance, car commuters ranged from cost-sensitive individuals to convenience-driven and disengaged habitual drivers. Public transport users included both financially motivated long-distance travelers and those prioritizing convenience, while cyclists emphasized practical facilities, with secure bike storage emerging as a key enabler.

Complementing this, a qualitative case study at ASML involved semi-structured interviews to explore the lived experiences of commuters. This revealed nuanced insights, such as the psychological role of commuting as a buffer between work and home, and the stress induced by traffic congestion and parking challenges. The “last mile” emerged as a critical barrier to public transport adoption, with inefficient final connections significantly reducing its appeal. Social influences and financial considerations also played context-dependent roles in shaping commuting behavior.

The findings underscore the need for mobility policies that reflect the segmented nature of commuter motivations rather than relying on demographic generalizations. Effective strategies must consider the entire door-to-door journey, as a single inconvenience can undermine broader policy goals. Moreover, perceived fairness and the availability of viable alternatives are essential for policy acceptance.

Based on these insights, the thesis recommends several actions for ASML and similar organizations. Addressing the last mile requires collaboration with municipalities and exploring remote office hubs. Financial incentives should be optimized, maintaining high-impact subsidies while recognizing the limited effect of smaller reimbursements. Parking management should shift toward fairness and demand regulation, with investments favoring sustainable alternatives like premium carpool spots. Finally, unlocking the potential of carpooling and other alternatives may require innovations such as a “Mobility Guarantee” to enhance flexibility. ... Read more

Over recent years, subjective well-being (SWB) has become a primary goal in urban planning, with research showing that the built environment can significantly influence residents’ well-being. This study focuses on the role of the subjective nature of aesthetic quality, which traditional segmentation-based computer vision approaches often fail to capture. To address this, we evaluate the Computer Vision-enriched Discrete Choice Model (CV-DCM) developed by Van Cranenburgh and Garrido-Valenzuela (2025), which uses a vision transformer and classifier to extract holistic visual features from Google Street View images and estimate continuous utility scores that reflect perceived visual quality, trained on stated trade-offs that people make between visual environments. We link these scores to life satisfaction and hedonic well-being measures from the Netherlands Mobility Panel (SWLS, 2020–2022; MHI-5, 2020) and analyze their relationships using Structural Equation Modeling (SEM), controlling for socio-demographic and built environment variables. Results show that PC5-level utility aligns more closely with life satisfaction than PC6, indicating that broader neighborhood context matters more than immediate street conditions. When non-linear age effects are modeled, a small but significant direct path from utility to life satisfaction emerges, whereas no significant association is found for hedonic well-being. Overall, the current explanatory power for SWB is modest and appears mainly driven by who lives where. Nevertheless, a perception-based computer vision model provides a scalable way that can quantify subjective visual quality, which could gain relevance when improved model fit is achieved by reducing variance in data collection or retraining the model on SWB-specific objectives. ... Read more

This research aims to provide a deeper understanding of determinants and outcomes of perceived accessibility by accounting for changes in perceived accessibility over time. This objective is addressed through a twofold empirical approach. First, changes in perceived accessibility and the factors driving these changes are analysed. Afterwards, the impacts of perceived accessibility on travel, activity participation, and well-being are empirically explored. The determinants and outcomes of perceived accessibility assessed in this thesis are identified through a literature review.

A longitudinal latent class analysis highlights various trajectories in perceived accessibility over time. Most trajectories in perceived accessibility (61% of the sample) remained relatively stable over time. Other trajectories indicate notable transitions, with two clusters (27% of the sample) reporting a decline and a single cluster (12% of the sample) reporting an increase between 2020 and 2023. The fixed and random effects regression models reveal that changes in the distance to the nearest amenities (in particular: supermarket, train station, secondary school) partly explain transitions in perceived accessibility. These effects are even more substantial for vulnerable groups. Still, other factors contribute to changes in perceived accessibility to a larger extent. These are changes in mobility tool ownership, household composition, and the number of times meeting online.

In the second part, pathways from perceived accessibility to life satisfaction have been explored using a twofold structural equation modelling (SEM) approach. First, the direct and indirect effects of perceived accessibility on life satisfaction have been systematically studied using 2020 data from MPN members in a cross-sectional SEM design. The analysis reveals that greater perceived accessibility is positively associated with more walking and activities such as grocery shopping and engaging in sports whilst negatively associated with health-related activities. Second, this thesis empirically explores a potential (bi)directional causal link between perceived accessibility and life satisfaction using a longitudinal SEM design. For this, a random-intercept cross-lagged panel model has been estimated using data from 2020, 2023, and 2024. This last analysis highlights that life satisfaction is more likely to positively change perceived accessibility than vice versa.

Based on these empirical findings, three main lessons for policy can be identified. First, changes in the physical environment are actively accounted for by individuals, especially for vulnerable groups. With the loss of amenities and, inherently, increasing distances to activity locations in the Netherlands, an important policy lesson is to identify the neighbourhoods where vulnerable groups live, monitor the corresponding trends in spatial accessibility, and address the decline in amenities in these neighbourhoods using policy designs. Second, providing individuals with the means to travel might mitigate the loss of a private ownership tool and, in turn, allow individuals to maintain their level of accessibility. In this respect, investments in shared mobility or public transport services such as trains, busses, and demand-responsive transport could be alternatives to mitigate private ownership losses. Last, this thesis did not find within-person effects from perceived accessibility to life satisfaction, suggesting that greater perceived accessibility may not be sufficient to enhance life satisfaction directly. Therefore, improvements in accessibility levels should most likely be part of a broader policy design that addresses (perceived) social inclusion and (subjective) well-being. ... Read more

Shared mobility has emerged as a promising strategy for promoting sustainable and efficient urban transport, yet its broader impacts—particularly when multiple shared modes are combined—remain insufficiently understood. This study investigates the relationship between shared mobility usage and private vehicle ownership and use in the Netherlands, drawing on a comprehensive national survey of shared mobility users. Employing binary logistic regression and latent class cluster analysis (LCCA), the research identifies six distinct user groups based on their shared mobility patterns, socio-demographic profiles, and spatial characteristics. The findings reveal that frequent use of station-based and community car sharing, as well as OV-fiets, is significantly associated with reductions in private car ownership and use. Other shared modes, such as free-floating car sharing, moped sharing, and shared e-bikes, also contribute to these reductions, though their effects are more modest and context-dependent. The study highlights that the most substantial decreases in car dependency occur among user groups with high engagement in integrated shared mobility services, particularly in urban areas with supportive infrastructure and policies. The study highlights the need for targeted policy interventions to improve accessibility, multimodal integration, and incentives for shared mobility, while also monitoring potential shifts away from public transport and active travel. This research advances understanding of the diverse effects of shared mobility in the Dutch context and offers evidence-based recommendations for maximising its sustainability benefits in urban transport systems. ... Read more

This thesis investigates whether everyday access to urban services can help explain patterns of democratic engagement in European cities. It treats accessibility not only as a functional condition of daily life but also as a spatial indicator of how present public institutions are in citizens’ neighbourhoods. The study asks to what extent proximity to essential services is associated with voter turnout and electoral volatility, interpreted respectively as symptoms of democratic disengagement and democratic discontent. Empirically, it combines walking Proximity Time measures from the Sony CSL accessibility dataset with electoral results for Paris, Berlin and Rome, mapped on a fine hexagonal grid. Rental costs are used as a proxy for local income in order to check whether accessibility effects simply mirror socio economic composition. The analysis relies on descriptive statistics, cartographic exploration and correlation matrices that link accessibility to turnout and volatility across different service domains, distinguishing between services provided or strongly regulated by the state and private amenities.

The results show that accessibility is related to democratic behaviour but in a context dependent way. In Paris and Rome, longer travel times to state related services are generally associated with lower turnout and, in Rome, with strongly concentrated belts of high volatility in peripheral areas. Berlin displays an opposite configuration in which more distant zones record slightly higher participation, a pattern that reflects its polycentric structure and relatively stable residential communities. Across all three cities, rental costs explain only a small fraction of the variation in accessibility, which suggests that planning legacies and urban morphology play an independent role. Overall, the findings provide suggestive but strictly correlational evidence. They indicate that accessibility can act as a territorial channel through which institutional presence is experienced, while also highlighting the need for longitudinal and quasi experimental research designs to establish whether changes in accessibility have causal effects on democratic engagement. ... Read more

Everyday use of ChatGPT spans study and work, but surveys and usage logs capture different parts of that behaviour. This thesis compares what people say in a short survey with what their anonymised ChatGPT export shows, without linking individuals. The design is simple and auditable: donors share platform-native logs, prompts are mapped to clear task families using compact example prototypes and all comparisons are made at the group level. To keep both sources directly comparable, the same frames are used on each side: intensity (how often and how long), timing (broad dayparts on one time base), input form (prompt-length bands) and task portfolio (main task families with concise subtasks). Analyses focus on full distributions and effect sizes rather than single averages. The core message is practical: self-reports give a workable signal for “how much,” while donation-based logs add detail on “how” people interact and “what” they use the tool for. Short, one-line, iterative or more technical exchanges are easy to miss in surveys, so using both sources together gives a more realistic picture for policy, training and procurement. The thesis closes with guidance on measuring and monitoring everyday use in organisations and education, with privacy and reproducibility built in. ... Read more

... Read more

The growing population in the Netherlands is straining the accessibility and liveability of cities. To address these issues, new mobility policies prioritize quiet, low-emission, and space-efficient modes of transport, creating a conflict with the characteristics of private fuel-engine cars, which are noisy, polluting, and space-inefficient. For example, a private car trip in Amsterdam requires about 95m² of public space, including 15m² for parking. Consequently, Dutch cities are considering measures to reduce car mobility, with parking regulations viewed as a potential tool to decrease car ownership and enhance urban liveability. However, there is a lack of evidence supporting the effectiveness of these policies, as car ownership research typically focuses on sociodemographic and household factors.

This study aims to fill this knowledge gap and assist municipalities in balancing car accessibility with urban liveability. It begins by defining urban parking availability, influenced not only by the number of parking spaces but also by various regulations. Two main types of parking are identified: private parking on premises and public on-street parking. Additionally, parking permits, which subsidize parking costs for residents, are considered.

To understand the influence of parking availability on household car ownership, a multinomial logistic regression (MNL) model was used. Data from 80,527 urban households over three years from the Dutch National Traffic Survey (ODiN) was analyzed. The study operationalized the number of on-street parking bays per household, parking places on premises per household, and maximum parking costs in the municipality.

The results showed that parking availability in a neighborhood does affect car ownership. Both types of parking were positively related to car ownership, while higher permit costs in a municipality decreased the likelihood of car ownership. However, household composition and income were the most influential factors. The number of driving licenses in a household, closely linked to the number of adults, and disposable income were the strongest predictors of car ownership.

The study also examined the combined influence of multiple parking availability dimensions. It compared car ownership probabilities in three neighborhoods for average households and predicted car ownership levels for a potential new neighborhood under different parking availability scenarios. The findings indicated significant increases in car ownership with higher parking availability, emphasizing the need for balanced parking strategies.

The relationship between parking availability and car ownership is crucial for policymakers, as it supports the formulation of parking regulations aimed at reducing car ownership. Municipalities are advised to develop parking policies that consider the identified dimensions of parking availability to optimize public space use.

Future research should focus on proving causality, improving spatial data, and exploring additional factors of car ownership and parking availability. The study's limitations include the inability to establish causality and the need for more reliable parking data. Further research should also consider attitudinal factors towards car ownership and investigate the relationship between parking availability and car ownership in rural areas. ... Read more