Introducing autonomous vehicles into urban mobility systems: A hybrid choice model approach

Abstract

Autonomous vehicles are expected to play a key role in the future of sustainable transportation, and their use is likely to expand across the world. As transportation systems continue to evolve, automated driving technologies and intelligent mobility solutions are becoming increasingly important. In this context, autonomous vehicles are anticipated to become a common part of urban mobility. However, the transition toward autonomous mobility also brings important challenges that require careful attention from researchers and policy makers. Beyond technological development, the way autonomous systems will interact with existing transport networks and user behaviour remains uncertain. While interest in autonomous mobility continues to grow, there is limited understanding of how individuals will respond to these systems, how their attitudes will evolve, and which factors will shape adoption and mode choice decisions. A key gap in the literature relates to the limited integration of behavioural, psychological, and preference-based perspectives when analysing the potential shift toward autonomous public transport. To address this gap, the first part of this thesis examines users' acceptance of autonomous public transport through an extended behavioural framework, identifying key determinants that influence intention to use these systems. Findings show that integrating satisfaction, loyalty, and trust into the framework significantly improves the explanation of intention to adopt autonomous public transport. Results also reveal that social influence, trust in automation, and loyalty to current public transport services play central roles in shaping user acceptance across different modes. As autonomous mobility approaches real-world implementation, it becomes increasingly important to understand how individuals will respond to emerging transport options and how their travel choices may change in an autonomous future. Previous research demonstrates that users' perceptions, expectations, and concerns, particularly regarding safety and reliability, play a crucial role in shaping acceptance and willingness to adopt new mobility technologies. In line with this need, the second component of this thesis investigates mode choice behaviour in the context of autonomous mobility, focusing on conventional bus users in Istanbul. This analysis explores how travel attributes, user characteristics, and latent attitudes influence preferences among conventional and autonomous transport alternatives. Understanding potential challenges and user attitudes toward autonomous mobility is highly important for transport planners and policy makers. As cities prepare for automated services, insight into the behavioural factors shaping adoption is essential. The results indicate that habitual inertia to conventional bus services remains a strong barrier to adopting autonomous alternatives, while perceived benefits and trust in automation substantially increase willingness to switch. Moreover, while the second study focuses on conventional bus users in Istanbul, expanding the analysis beyond a single-city context strengthens the ability to understand and compare autonomous mobility adoption across diverse urban environments. To provide a broader perspective, the third component applies a similar analytical framework to London, a city characterised by one of the most mature and multimodal transport networks in the world. Unlike Istanbul, where the emphasis is on transitioning from conventional buses to autonomous bus and ride-hailing services, the London study incorporates a wider array of modes including walking, cycling, private car, taxi, bus, and rail. This approach enables a more comprehensive assessment of how autonomous services may interact with established public transport and active mobility options. The findings highlight that attitude toward cars and public transport strongly shape acceptance of autonomous services, with trust in automation emerging as a critical enabler in technology-rich, multimodal settings. Additionally, individuals who already use shared or public transport show greater openness to autonomous shared modes, whereas strong preference for private cars continues to challenge the shift toward automated shared mobility. From a methodological perspective, this thesis presents an integrated analytical framework that combines behavioural modelling, latent attitudinal analysis, and discrete choice modelling to examine the future adoption of autonomous mobility solutions. While each of these approaches is well established in transport behaviour research, their combined application in the context of autonomous public transport and shared autonomous services offers a novel contribution. The framework adopts both traditional econometric structures and hybrid choice modelling techniques, allowing attitudinal factors such as trust, satisfaction, and technology perception to be incorporated alongside observable travel attributes. This multidimensional structure enhances the explanatory power of mode choice models and provides a more realistic representation of the cognitive and emotional factors affecting autonomous vehicle adoption. Furthermore, the methodological approach is designed to be transferable and can be applied to different urban contexts, offering a robust basis for cross-city comparison of behavioural responses to autonomous mobility. Overall, this thesis underscores the importance of adopting human-centred analytical frameworks in future mobility research, particularly during the early transition toward automation. By integrating psychological constructs with travel behaviour modelling, the study advances the understanding of how attitudes, habits, and perceived risks shape acceptance of autonomous transport systems. The methodological innovations, empirical insights drawn from Istanbul and London case studies, and policy-relevant findings make this research a valuable reference for academics, practitioners, and decision makers. In doing so, it contributes to the theoretical development of behavioural transport modelling, enriches the empirical evidence base on autonomous mobility adoption, and provides practical guidance for designing user-oriented, equitable, and sustainable autonomous public transport strategies.