Travel Genie: An Intelligent End-to-End Travel and Itinerary Planning System

Abstract

Travel itinerary planning is a complex optimization problem that requires balancing multiple factors such as budget, trip duration, accommodation preferences, attraction categories, transportation accessibility and user interests. While existing travel planning systems can generate personalized itineraries, they often lack a comprehensive optimization framework that simultaneously considers geographical proximity, user preferences. To address this challenge, this paper presents Travel Genie, an intelligent travel and itinerary system designed to generate intelligent and personalized travel itineraries for tourists visiting London. The proposed framework integrates clustering and optimization techniques including K-Means Clustering, Constraint Satisfaction Problem (CSP), Mixed Integer Linear Programming (MILP) and Genetic Algorithm (GA) to generate feasible and personalized travel itineraries. User preferences including trip type, budget, group size, trip duration, preferred attraction categories and accommodation requirements are collected through an interactive interface. K-Means clustering groups geographically similar Points of Interest (POIs) to reduce travel complexity, while CSP filters infeasible travel options based on user-defined constraints. MILP then selects an optimal set of attractions that maximizes itinerary quality while satisfying all constraints. Finally, the Genetic Algorithm enhances itinerary diversity and minimizes unnecessary travel between attraction clusters. The generated itineraries are organized into daily schedules with recommended accommodations, nearby transportation information and time-slot allocation for each Point of Interest through a user-friendly Streamlit interface. Experimental evaluation demonstrates the effectiveness of the proposed framework in generating feasible and personalized itineraries with improved clustering quality, high constraint satisfaction and efficient itinerary optimization. The proposed approach provides an effective decision-support system for automated travel planning and enhances the overall travel experience through intelligent itinerary generation.

Introduction

The paper presents Travel Genie, an intelligent travel itinerary planning system designed to generate personalized and optimized travel plans for tourists visiting London. The system addresses the limitations of traditional travel planning methods, which are often time-consuming and inefficient due to manual searching, accommodation selection, transportation planning, and scheduling difficulties.

Travel Genie integrates Artificial Intelligence, Machine Learning, and optimization techniques to create user-specific itineraries based on preferences such as budget, trip duration, group size, accommodation requirements, trip type, and preferred attractions. The proposed framework combines K-Means Clustering for grouping geographically related Points of Interest (POIs), Constraint Satisfaction Problem (CSP) for filtering infeasible travel options, Mixed Integer Linear Programming (MILP) for selecting optimal attractions, and Genetic Algorithm (GA) for improving itinerary diversity and travel efficiency.

The system uses three datasets: POIs, hotels, and transportation facilities in London. After preprocessing, POIs are clustered geographically, and user constraints are applied to identify feasible options. MILP optimizes attraction selection, while GA further enhances the final itinerary by balancing attraction diversity, quality, and travel efficiency. The generated output includes day-wise schedules, recommended accommodations, transportation information, and time allocations for each attraction through a Streamlit-based user interface.

The literature review highlights that previous travel recommendation systems mainly focus on individual POI recommendations or limited optimization objectives. Unlike existing approaches, Travel Genie provides a unified framework that combines geographical clustering, constraint handling, mathematical optimization, and evolutionary refinement.

The system performance is evaluated using Silhouette Score, Preferred Category Coverage, and Constraint Satisfaction Rate. The obtained results show a Silhouette Score of 0.461, Preferred Category Coverage of 84%, and Constraint Satisfaction Rate of 95%, demonstrating effective clustering, strong personalization, and successful handling of user-defined constraints.

Conclusion

This paper presented Travel Genie, an intelligent travel and itinerary planning system that generates personalized and optimized travel itineraries by integrating user preferences, accommodation recommendations, transportation information, and multiple computational techniques within a unified framework. The proposed system combines K-Means Clustering for geographically grouping Points of Interest (POIs), a Constraint Satisfaction Problem (CSP) for identifying feasible travel options, Mixed Integer Linear Programming (MILP) for optimal attraction selection, and a Genetic Algorithm (GA) for refining itinerary quality and attraction diversity. Experimental evaluation using the Silhouette Score, Preferred Category Coverage and Constraint Satisfaction Rate demonstrates the effectiveness of the proposed framework in generating feasible, optimized, and diverse travel itineraries. The results indicate that the system effectively reduces manual travel planning effort while providing personalized, practical, and efficient travel recommendations. Future work will focus on extending the proposed framework by incorporating real-time traffic, weather, hotel availability, and transportation data to support dynamic itinerary generation. Integration with live hotel booking and flight reservation APIs can further enhance the practicality of the system by enabling real-time cost estimation and booking facilities. Additionally, the framework can be extended to support multi-city and international trip planning, multilingual interaction, and offline navigation capabilities. These enhancements would improve the scalability, adaptability, and real-world applicability of the proposed travel planning system

References

[1] Chang, C.-W., Tsai, C.-Y., Yao, L., Kuo, R. J., and Tsai, C.-Y., \"Personalized Travel Itinerary Recommendation Enhancing by User Interests and Point-of-Interest Characteristics,\" Information Technology & Tourism, vol. 27, pp. 649–682, 2025, doi: 10.1007/s40558-025-00318-2. [2] H. Li, M. Wang, J. Zhang, T. Shi, and A. Khamis, \"A Contextual Multi-armed Bandit Approach to Personalized Trip Itinerary Planning,\" in Proc. 2024 IEEE International Conference on Smart Mobility (SM), pp. 55–60, 2024. doi: 10.1109/SM63044.2024.10733530. [3] N. S. Prabha, G. Kalaiarasi, V. Nivash, M. Selvi, and R. Yogitha, \"Personalized Travel Itinerary Generator System,\" in Proc. 2024 International Conference on Intelligent Systems for Cybersecurity (ISCS), 2024. doi: 10.1109/ISCS61804.2024.10581212. [4] T. Ho and A. Lim, \"POIBERT: A Transformer-based Model for the Tour Recommendation Problem,\" in Proc. IEEE International Conference on Big Data (Big Data), 2022, pp. 4935–4944. doi: 10.1109/BigData55660.2022.10020877. [5] D. Jawale, P. Mishra, P. Kamble, D. Mhatre, S. Jadhav, and L. Singh, \"AI-Optimized Itinerary Design: Transforming the Future of Travel Planning,\" Journal of Artificial Intelligence Research & Advances, vol. 12, no. 2, pp. 74–82, Apr. 2025. doi: 10.37591/JoAIRA. [6] C. Almira and N. U. Maulidevi, \"Travel Itinerary Recommendation for Real World Point of Interests Using Iterated Local Search,\" in Proc. 2019 International Conference of Advanced Informatics: Concepts, Theory and Applications (ICAICTA), 2019, pp. 1–6, doi: 10.1109/ICAICTA.2019.8904339. [7] Chen, S. Wu, J. Zhou, and A. K. H. Tung, \"Automatic Itinerary Planning for Traveling Services,\" IEEE Transactions on Knowledge and Data Engineering, vol. 26, no. 3, pp. 514–527, Mar. 2014. doi: 10.1109/TKDE.2013.46. [8] S. Dhote, N. Gayakwad, P. Buwade, R. Bedre, and P. S. Sisodiya, \"AI-Powered Travel Planner: Smart Trip Optimization,\" International Journal of Leading Research Publication (IJLRP), vol. 6, no. 3, Mar. 2025, E-ISSN: 2582-8010. [9] A. Gawade, S. Jadhav, P. Patil, R. Khule, and P. K. Mishra, \"AI-Powered Personalized Travel Itinerary Planning Using Real-Time Data,\" International Journal of Scientific Advances and Technology (IJSAT), vol. 2, no. 4, pp. 1–7, 2025. [10] L. Chen, L. Zhang, S. Cao, Z. Wu, and J. Cao, \"Personalized Itinerary Recommendation: Deep and Collaborative Learning with Textual Information,\" Expert Systems with Applications, vol. 144, Art. no. 113070, 2020. doi: 10.1016/j.eswa.2019.113070.

Copyright

Copyright © 2026 Gollapalli Akshitha, Dr. V. Uma Rani. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.