Abstract: As international tourism expands and synthetic intelligence era advances, smart travel planning offerings have emerged as a significant research awareness. Within dynamic actual- global journey scenarios with multi-dimensional constraints, offerings that guide users in automatically developing practical and customized journey itineraries must cope with three key objectives: Rationality, Comprehensiveness, and Personalization. However, existing systems with rule-primarily based combos or LLM-primarily based planning strategies conflict to fully fulfill these criteria. To triumph over the demanding situations, we introduce “TravelAgent”, a tour making plans system powered with the aid of massive language fashions (LLMs) designed to offer affordable, comprehensive, and customized travel itineraries grounded in dynamic situations. TravelAgent incorporates four modules: Tool-utilization, Recommendation, Planning, and Memory Module. We evaluate TravelAgent’s performance with human and simulated customers, demonstrating its typical effectiveness in three criteria and confirming the accuracy of customized tips.

| DOI: 10.17148/IARJSET.2025.125334