A fuzzy-integrated hybrid gradient-boosted AI system for resilient, safe, and sustainable heritage tourism planning

Heritage tourism trip planning is essential for ensuring visitor satisfaction, preserving cultural sites, and supporting local economies, especially in regions like northern Thailand. Existing methodologies often fail to comprehensively address sustainability, safety, resilience, economic impact, and cultural preservation. To address this gap, this research presents the Hybrid Gradient-Boosted Artificial Intelligence System (GBAMIS). GBAMIS employs a mixed approach for initial solution construction and uses Gradient Boosted Machines (GBM) to optimize performance within the Artificial Multiple Intelligence System (AMIS) framework. Additionally, 12 novel heuristic-based improvement equations are introduced. Data from 125 attractions and feedback from 450 tourists in northern Thailand were used to validate the system. Key indices, including Sustainability, Safety, Resilience, Economic Impact, and Cultural Preservation, were assessed. GBAMIS outperformed existing methods, with hypervolume improvements of 18.75%, 14.46%, and 6.74% over AMIS, GRASP-VND, and MRL-AMIS, respectively. It also showed significant gains in Pareto optimal solutions, improving by 19.48%, 15%, and 6.98%. Furthermore, GBAMIS demonstrated strong performance under uncertainty through fuzzy membership management, handling varying values with consistent results. This adaptability makes it suitable for dynamically changing environments, offering a novel solution for heritage tourism planning. The findings highlight GBAMIS’s potential to enhance cultural site preservation, visitor experiences, and economic support for local communities, contributing to advancements in intelligent tourism management systems.