This study investigates the growing role of wearable technologies in cruise tourism, focusing on their effects on passengers’ individual performance and cruise vacation wellbeing. Using the belief-action-outcome (BAO) framework, the research examines how factors such as perceived convenience, privacy concerns, social influence, and perceived benefits shape passengers’ behavioral intention and actual use of wearable sensors onboard. A structured survey was conducted among 301 cruise passengers. The data were analyzed using partial least squares structural equation modeling (PLS-SEM) to assess relationships among key variables, including intention to use, self-reported usage frequency, performance outcomes, and cruise vacation wellbeing. The study also explores moderating effects of social influence and passenger loyalty. The analysis indicates that both intentions to use and self-reported usage frequency of wearable devices positively influence passengers’ onboard performance and cruise vacation wellbeing. Social influence was found to moderate the impact of privacy concerns on intention, while loyalty moderated the link between usage and wellbeing. These results highlight the dual role of personal and social factors in wearable tech adoption. This research expands the BAO framework by applying it in a smart tourism setting, specifically within cruise travel. It offers new insights into how consumer-facing technologies can enhance passenger experience, guiding cruise operators in leveraging wearables to optimize service delivery and support passengers’ individual performance and broader perceived cruise vacation wellbeing.
Artificial intelligence (AI) is reshaping hospitality and tourism, yet adoption among small and medium-sized enterprises (SMEs) remains uneven and cannot be explained by cost, readiness, or perceived usefulness alone. This study presents a qualitative meta-synthesis of ten qualitative and mixed-methods studies with extractable qualitative findings, published or available online-first between January 2020 and April 2026, to explain how AI-enabled and AI-adjacent smart technologies are adopted, resisted, and may contribute to adaptive resilience in hospitality and tourism SMEs. Using procedures informed by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses, the Critical Appraisal Skills Programme, a 6Ws extraction protocol, and thematic synthesis, the study integrates evidence through a combined Technology–Organization–Environment and Sociotechnical Systems lens. The synthesis identifies three resistance filters—identity conflict, structural fragility, and ethical trust—and four strategic enablers: hybridization, platform/intermediary scaffolding, twin-transition alignment, and assurance practices. The findings show that adoption is not a linear technological upgrade but a sociotechnical alignment process. AI-enabled and AI-adjacent smart technology adoption becomes more acceptable when technologies augment rather than replace human service, when intermediaries compensate for internal capability gaps, when digital tools support sustainability outcomes, and when trust is reinforced through transparency, cybersecurity, fairness, human oversight, and accountability. The study contributes by reframing resistance as partly identity-protective, explaining how structural adoption conditions are filtered through service norms and human roles, and conceptualizing ethical trust as an assurance requirement. Practical implications include staged back-of-house-first adoption, careful platform partnership management, twin-transition-oriented investment, and assurance-by-design for SME-facing AI systems. This qualitative meta-synthesis explains uneven AI-enabled and AI-adjacent smart technology adoption among hospitality and tourism SMEs. Drawing on ten empirical studies, it shows that adoption is shaped by three resistance filters—identity conflict, structural fragility, and ethical trust—and enabled through hybridization, platform/intermediary scaffolding, twin-transition alignment, and assurance practices. The SME–AI Adaptive Resilience Framework reframes adoption as a sociotechnical alignment process linking efficiency, authenticity, trust, sustainability, and adaptive resilience.
Information and communication technologies have transformed the ways in which tourists access, interpret and share information about urban destinations. This study examines the relationship between multilingual Wikipedia visibility and Instagram hashtag intensity across tourist attractions in Seville. To do so, it develops and validates an AI-based indicator of attraction prominence derived from twelve Wikipedia language editions and combines linear mixed-effects modelling, Random Forest regression and cluster-specific analyses. The results show that Wikipedia-derived visibility indicators are significantly associated with Instagram hashtag intensity, particularly among language groups that are geographically and culturally more distant from Spain. The findings suggest that Wikipedia operates as a multilingual knowledge infrastructure within the broader digital information ecosystem of urban tourism. The study contributes methodologically by demonstrating how large language models can be used to extract comparable indicators from multilingual textual sources, provided that validation procedures are applied. The findings also highlight the importance of linguistic and cultural heterogeneity when analysing tourism-related digital traces.
Agentic AI (aAI) is moving tourism beyond recommendation and search toward autonomous decision and action. When algorithms select, sequence, and even book, the centre of decision-making shifts. This viewpoint clarifies how aAI differs from generative AI, and reflects on the consequences for consumer behaviour, market visibility, and managerial practice. We argue that aAI can either homogenise experiences around popular options or reveal hidden diversity depending on design choices and governance. For managers, marketing to the machine, data strategy, and platform dependence become central, while human contact may return as a differentiator. We also outline a research agenda on transparency, redistribution of visitor flows, and cross-border regulation. Rather than celebrating or dismissing aAI, we call for a critical and practical approach that keeps efficiency, fairness, and authentic choice in balance.
Understanding fine-scale tourist mobility is essential for destination planning and operational management, including time-windowed staffing, crowd mitigation around hubs and corridors, and evaluation of place-based interventions. However, high-frequency GPS trajectories are noisy, irregularly sampled, and weakly linked to the place information needed for such decisions. This study develops a probabilistic event-based framework that converts GPS trajectories into ordered stay events, with each event represented by a normalized MID10 category-compatibility vector derived from nearby POIs. To improve reproducibility, distance-weighted compatibility estimates are averaged across candidate-set sizes, preserving event-to-place ambiguity without relying on a single hard match or stochastic candidate set. POI semantics therefore represents category compatibility around an observed stay rather than a verified facility visit or activity purpose. Using one month of GPS trajectories from Hakone, Japan (November 2021), we construct compatibility-weighted event sequences and estimate temporal profiles, category transitions, and dwell distributions. We then formulate a semi-Markov simulator with time-conditioned transitions, category-dependent dwell durations, and empirically constrained spatial POI instantiation. Baseline validation shows close agreement in hourly Expected Stays (ES) and Expected Dwell Minutes (EDM) (median r = 0.994 and 0.986), while spatial and movement diagnostics identify remaining discrepancies in inter-event distance and implied speed. Under the no-boost specification, type-switch, addition, relocation, and removal scenarios produce small, localized redistributions around intervention POIs; these scenario changes are interpreted under fixed model components rather than as evidence of causal tourist responses.
