Design and Implementation of Artificial Intelligence System for Halal Travel Route Recommendation

The rapid growth of halal tourism has created demand for intelligent systems that can generate travel recommendations aligned with both logistical efficiency and Shariah principles. This study presents the design and implementation of the Halal Phuket system, an AI-driven platform developed to support Muslim-friendly tourism in Phuket Province, Thailand. The system is structured into three modules: (1) the Data Management Module, which ensures accurate and up-to-date information on mosques, halal restaurants, hotels, and attractions; (2) the Tourist Application Module, built with Flutter for cross-platform usability, which provides search, filtering, GPS-based recommendations, and real-time prayer-time notifications; and (3) the Route Planning Module, which integrates Genetic Algorithms (GA), fuzzy logic, and generative AI to compute optimized halal travel itineraries enriched with descriptive narratives. System testing confirmed the effectiveness of the approach. CRUD operations achieved a 100% success rate with negligible errors, and search and filtering accuracy exceeded 92% across multiple criteria. The GA optimization improved route fitness values by 18–25% compared with baseline itineraries, while consistently maintaining high prayer-time satisfaction scores (SMS ≥ 0.85). Real-time GPS features reliably detected nearby halal facilities within a 200-meter margin, and generative AI route descriptions received high ratings for readability (4.6/5) and cultural alignment (4.7/5) from subject matter experts (SMEs). The findings demonstrate that Halal Phuket successfully integrates AI optimization, cultural governance, and user-centric design to deliver practical and Shariah-compliant travel planning. This research contributes a novel framework for applying AI in halal tourism and introduces the prayer-time satisfaction metric as a unique measure of religious adherence.