Crisis Management Days

Artificial Intelligence or Human Decision-making: A Study on Crisis Response Following an Earthquake

Igor Milić

University of Applied Sciences Velika Gorica

Krunoslav Bilić

Zagreb University of Applied Sciences

Hrvoje Janeš

University of Applied Sciences Velika Gorica

Keywords: Earthquake, decision making, crises coordination, AI analysis

---

Abstract

Decision-making in earthquake crises requires rapid coordination, informed analysis, and ethical responsibility. This study explores the comparative strengths of artificial intelligence (AI) and human judgment in such contexts through a simulated earthquake response scenario. The simulation was conducted as part of the 2022 Civil Protection Exercise Plan of the City of Velika Gorica, supported by Velika Gorica University of Applied Sciences and the Directorate of Civil Protection, with the goal of strengthening the city’s disaster response readiness. The case examines how AI could contribute to operational efficiency and pattern recognition, while also recognizing the enduring importance of human expertise in ethical decision-making and adaptive leadership. Findings indicate that a hybrid model combining AI-driven tools with human oversight offers the most resilient framework for crisis response. The study underscores the need for continued development of integrated decision-support systems grounded in practical exercises and cross-sector collaboration.

---

References

Ahmad, M. (2025). An enhanced deep reinforcement learning approach for humanitarian logistics. Engineering Applications of Artificial Intelligence.

Boin, A., ’t Hart, P., Stern, E., & Sundelius, B. (2005). The politics of crisis management: Public leadership under pressure. Cambridge, UK: Cambridge University Press.

Carvalho, P. (2025). A collaborative taxonomy of social media indicators for disasters. PLOS ONE.

Cheng, M. Y. (2024). Computer vision-based post-earthquake inspections for buildings. Automation in Construction.

Comfort, L., Ko, K., & Zagorecki, A. (2004). Coordination in rapidly evolving disaster response systems: The role of information. American Behavioral Scientist, 48, 295–313. https://doi.org/10.1177/0002764204268987

Coombs, W. T. (2015). Ongoing crisis communication: Planning, managing, and responding (4th ed.). Thousand Oaks, CA: Sage.

He, C. (2025). Social media analytics for disaster response. Applied Sciences.

Khallouli, W. (2025). Integrated deep learning for emergency tweet identification (BERT + rules). Social Network Analysis and Mining.

Kızılay, F. (2024). Evaluating fine-tuned deep learning models for real-time building damage detection from UAV imagery. Discover Artificial Intelligence.

Kumar, S., & Havey, T. (2013). Before and after disaster strikes: A relief supply chain decision support framework. International Journal of Production Economics, 145(2), 613–629. https://doi.org/10.1016/j.ijpe.2013.05.016

Organisation for Economic Co-operation and Development (OECD). (2025). Using AI to measure disaster damage costs: Methodology and the example of the 2018 Sulawesi earthquake. Paris, France: OECD Publishing.

Özcan, Ş. (2021). Effectiveness in crisis management. Ankara, Turkey: Iksad Publishing House.

Sidahmed, E. (2024). Strategic support systems for crisis management: A literature review. Foundations and Trends in Information Systems.

Turoff, M., Chumer, M., Van de Walle, B., & Yao, X. (2004). The design of a dynamic emergency response management information system (DERMIS). Journal of Information Technology Theory and Application, 5(4), 1–35.

Youngblood, S. (2010). Ongoing crisis communication: Planning, managing, and responding, 2nd edition (Coombs, W. T.) and Handbook of Risk and Crisis Communication (Heath, R. L., & O’Hair, H. D., Eds.) [Book reviews]. IEEE Transactions on Professional Communication, 53, 174–178. https://doi.org/10.1109/TPC.2010.2046099

Yu, L. (2021). Reinforcement learning approach for resource allocation in humanitarian logistics. Expert Systems with Applications.

Zheng, Z. (2024). Towards transferable building damage assessment via domain-adaptive change detection. Remote Sensing of Environment.