Predicting Passenger Preferences: An AI-Driven Framework for Personalized Airport Lobby Experiences

The modernization of airport facilities and services has become increasingly focused on enhancing passenger experiences. Among these experiences, personalized airport lobby experiences stand out as a key area for improvement. By tailoring services and amenities to individual preferences, airports can create more engaging and satisfying experiences for travelers. To achieve this level of personalization, it is imperative to accurately predict passenger preferences during their time in airport lobbies. In this research work, we present a novel AI-driven framework, named HybridGNN, designed for predicting passenger preferences to enhance personalized airport lobby experiences. The proposed framework integrates a hybrid recommender system with graph neural networks (GNNs), leveraging the strengths of both approaches to provide accurate and personalized recommendations. Through comprehensive simulation analysis, HybridGNN is evaluated against existing algorithms using simulation metrics such as accuracy, mean squared error (MSE), training time, and precision. Results indicate that HybridGNN consistently outperforms existing algorithms across all metrics, demonstrating its effectiveness in predicting passenger preferences. This research contributes to the advancement of personalized airport experiences and underscores the potential of AI-driven frameworks in enhancing passenger satisfaction and engagement.