كنفرانس بين المللی مهندسی صنايع و سيستم ها

صفحه اصلی / دوازدهمین كنفرانس بين المللی مهندسی صنايع و سيستم ها

Agentic AI Framework for Adaptive Visitor-Flow Management in Smart Tourism Destinations

نویسندگان :

Fatemeh Ghorehzadeh¹ Reza Kia² Ata Allah Taleizadeh³

1- University of Tehran 2- Sultan Qaboos University 3- University of Tehran

کلمات کلیدی :

Agentic AI،Reinforcement learning،Q-learning،Smart tourism destination،Visitor-flow management،Pricing

چکیده :

Smart tourism destinations need intelligent decision-making mechanisms for addressing uncertainty in visitor demands, congestion, satisfaction and sustainability issues. The present paper suggests a simple Agentic Artificial Intelligence (AI) mechanism for smart visitor-flow management based on tabular Q-learning algorithm. In the suggested model, the destination manager is considered an autonomous agent that observes the environment of tourism and makes actions of promotion for controlling the visitor flow. An agent gains experience by interacting with the simulated tourism environment and gets rewards depending on revenue, visitor satisfaction, decrease of congestion and promotion costs. The proposed strategy is tested against random and heuristic benchmark strategies. According to simulation results, the Q-learning agent outperforms other agents by making balanced decisions which result in the decrease of congestion and improvement of visitors' satisfaction. The study demonstrates the potential of Agentic AI for supporting intelligent, adaptive, and sustainable tourism destination management.