Journal "Software Engineering"
a journal on theoretical and applied science and technology
ISSN 2220-3397

Issue N9 2026 year

DOI: 10.17587/prin.17.520-532
Intelligent Multi-Agent RAG System for Flood Emergency Public Information
T. Yu. Chernysheva, Cand. (Eng.), Associate Professor, t.y.chernysheva@utmn.ru, R. S. Shiklyaev, Student, stud0000244870@utmn.ru, D. K. Drachev, Student, stud0000273054@utmn.ru, University of Tyumen, Tyumen, 625003, Russian Federation
Corresponding author: Tatiana Yu. Chernysheva, Cand. (Eng.), Associate Professor, University of Tyumen, Tyumen, 625003, Russian Federation, E-mail: t.y.chernysheva@utmn.ru
Received on December 11, 2025
Accepted on March 17, 2026

The article presents the development and evaluation of an intelligent information system designed to support rapid decision-making and public communication during flood emergencies. The system is based on a Retrieval-Augmented Generation (RAG) approach and integrates official sources, domain knowledge and user interaction through an automated messaging interface. The goal of the research is to provide accurate, structured and per­sonalized recommendations for citizens by combining document retrieval with natural language generation. The architecture of the system incorporates modules for data acquisition, semantic indexing, query routing, and response generation, supported by a formal mathematical description of vector search and graph-based state transitions. A multi-agent design is applied to separate functional roles: operational guidance, legal information, and general inquiries. Each agent processes incoming requests within its domain, ensuring correct classification of user queries and structured output. The system retrieves relevant information dynamically, generates responses and formats them according to predefined templates, which contributes to reducing errors and ambiguity. A series of experi­ments was conducted on a representative set of user scenarios, including evacuation instructions, legal regulations and general awareness questions. Quantitative evaluation demonstrated high accuracy of responses and low er­ror rates, while response time remained suitable for real-time use. Qualitative examples confirmed consistency of output structure and relevance to user requests. The results show that the proposed system improves the clarity, speed, and reliability of emergency information delivery and may be applied to other risk management contexts requiring personalized, source-based recommendations, while also outlining prospects for its future integration with simulation-based evacuation models.

Keywords: RAG, LLM, multi-agent architecture, floods, civil protection, information systems, Telegram bot, se­mantic search, emergency situations
pp. 520—532
For citation:
Chernysheva T. Yu., Shiklyaev R. S., Drachev D. K. Intelligent Multi-Agent RAG System for Flood Emergency Public Information, Programmnaya Ingeneria, 2026, vol. 17, no. 9, pp. 520—532. DOI: 10.17587/prin.17.520-532. (in Russian).
This work was supported by the Ministry of Science and Higher Education of the Russian Federation within the project "Fundamental problems of development methodology and related legal and ethical regulation in the field of application of artificial intelligence systems and models" (FEWZ-2024-0052).
References:
  1. Karapuzikov A. A., Muraev N. P. Application of Innovative Technical Means in Conditions of Flooding of Territories, Pravovaia Informatika, 2025, no. 3, pp. 90—95. DOI: 10.24412/1994-1404­2025-3-00-09 (in Russian).
  2. Klesel M., Wittmann H. F. Retrieval-Augmented Generation (RAG), Business & Information Systems Engineering, 2025, vol. 67, pp. 551—561. DOI: 10.1007/s12599-025-00945-3.
  3. Jiao J., Park J., Xu Y. et al. SafeMate: A Modular RAG-Based Agent for Context-Aware Emergency Guidance, arXiv: pre-print, 2025. DOI: 10.48550/arXiv.2505.02306.
  4. Xu S., Yan Z., Dai C., Wu F. MEGA-RAG: a retrieval-augmented generation framework with multi-evidence guided an­swer refinement for mitigating hallucinations of LLMs in public health, Frontiers in Public Health, 2025, vol. 13. DOI: 10.3389/fpubh.2025.1635381
  5. Xie Y., Jiang B., Mallick T. et al. WildfireGPT: Tailored large language model for wildfire analysis, arXiv: preprint, 2024. DOI: 10.48550/arXiv.2402.07877.
  6. Buryi A. S., Tsaplina O. S. Agent-Based Models for De­cision-Making Using Generative Artificial Intelligence, Pravovaia Informatika, 2025, no. 1, pp. 101—109. DOI: 10.24412/1994-1404­2025-1-101-109 (in Russian).
  7. Helbing D., Molnar P. Social force model for pedestrian dynamics, Physical Review E, 1995, vol. 51, no. 5, pp. 4282—4286. DOI: 10.1103/PhysRevE.51.4282.
  8. Makarov V. L., Bakhtizin A. R., Beklaryan G. L., Akopov A. S. Development of Software Framework for Large-Scale Agent-Based Modeling of Complex Social Systems, Programmnaya Ingeneria, 2019, vol. 10, no. 4, pp. 167—177. DOI: 10.17587/prin.10.167-177 (in Russian).
  9. Akopov A. S., Beklaryan L. A. Agent model of crowd behavior in emergency situations, Avtomatika i Telemekhanika, 2015, no. 10, pp. 131—143 (in Russian).
  10. James A., Trovati M., Bolton S. Retrieval-Augmented Generation to Generate Knowledge Assets and Creation of Ac­tion Drivers, Applied Sciences, 2025, vol. 15, no. 11, article 6247. DOI: 10.3390/app15116247.
  11. Moser D., Bender M., Sariyar M. A Pipeline for Automat­ing Emergency Medicine Documentation Using LLMs with Re­trieval-Augmented Text Generation, Applied Science, 2025, vol. 39, no. 1, article 2519169. DOI: 10.1080/08839514.2025.2519169.
  12. Lewis P., Perez E., Piktus A. et al. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks, Advances in Neural Information Processing Systems, 2020, no. 33, pp. 9459—9474. DOI: 10.48550/arXiv.2005.11401.
  13. Kashirina I. L., Osipov I. R., Yakovlev V. A. Develop­ment and Evaluation of a Rag System for Analysis of Semantic Relations of Specialised Texts, Vestnik VGU. Seriia: Sistemnyi analiz i informatsionnye tekhnologii, 2025, no. 2, pp. 114—126 (in Russian).
  14. Gao Y., Xiong Y., Gao X. et al. Retrieval-Augmented Generation for Large Language Models: A Survey, arXiv: preprint, 2023. DOI: 10.48550/arXiv.2312.10997.
  15. Barnett S., Kurniawan S., Thudumu S. et al. Seven Failure Points When Engineering a Retrieval Augmented Generation System, Proceedings of the IEEE/ACM 3rd International Conference on AI Engineering-Software Engineering for AI, Lisbon, Portugal. April 2024, pp. 194—199.
  16. Guerraoui R., Kermarrec A., Petrescu D. et al. Efficient Federated Search for Retrieval-Augmented Generation, Proceedings of the 5th Workshop on Machine Learning and Systems, 2025, pp. 74—81.
  17. Gao Y., Xiong Y., Wang M. et al. Modular RAG: Transforming RAG Systems into LEGO-Like Reconfigurable Frameworks, arXiv: preprint. 2024. DOI: 10.48550/arXiv.2407.21059.