Recognizing AI-Based Personalization and Real-Time Crowd Management Through Smart Tourism Applications

At present, the rapid advancement of smart tourism applications plays a crucial role in promoting personalized experiences and helping efficient crowd management. The main objective of the research was to investigate recognition of AI-based models: Structural Equation Modeling, TOPSIS approach, and conceptual mapping in evaluation of personalization of real-time crowd management, respectively. For this purpose, data of three past tourism seasons and multiple operational indicators, namely, visitor flow, service quality, satisfaction level, waiting time, resource allocation (transportation), mobile interaction, staff responsiveness, safety perception, booking efficiency, and environmental comfort were considered in the analytical framework. Instead of using traditional benchmarks, matched decision matrices in this study can capture more dynamic features for better evaluation of smart tourism performance, although this article proposes only the preliminary process of matched evaluation. This research collected survey responses in the recent period (2022 to 2024) and used the SEM model and TOPSIS model to analyze the structural relationships between the applications and various service indicators in tourism. The key method used for designing matched evaluation is the conceptual mapping scheme, which has never been reported in the literature for AI-based matched assessment for smart tourism. Therefore, the framework is useful in environments with high variability, as it reduces evaluation bias when uncertainty level is significant, by prioritizing the least stable criteria. Because of the limited level of generalization in this review, further empirical validations are still needed to confirm robustness.