AI-driven decision-making for water resource planning and hazard mitigation using automated multi-agents

Likith Kadiyala; Ramteja Sajja; Yusuf Sermet; Marian Muste; Ibrahim Demir

doi:10.2166/hydro.2025.042

Back

AI-driven decision-making for water resource planning and hazard mitigation using automated multi-agents

Journal article

Open access

Peer reviewed

AI-driven decision-making for water resource planning and hazard mitigation using automated multi-agents

Likith Kadiyala, Ramteja Sajja, Yusuf Sermet, Marian Muste and Ibrahim Demir

Journal of hydroinformatics, Vol.27(7), pp.1217-1237

06/23/2025

DOI: 10.2166/hydro.2025.042

Appears in UI Libraries Support Open Access

Files and links (1)

url

https://doi.org/10.2166/hydro.2025.042View

Published (Version of record) Open Access

Abstract

This project simulates the multi-hazard tournament (MHT) framework, a decision-support system designed for the U.S. Army Corps of Engineers, using artificial intelligence (AI) agents to enhance decision-making processes for flood mitigation and water resource management. Traditional hydrological models often overlook social dynamics and community preferences crucial for sustainable implementation. The objective of the framework is to develop optimal strategies for protecting water resources, habitats, and communities within a defined budget. The simulation integrates AutoGen for managing multi-agent interactions and DarkIdol-Llama-3.1-8B, an advanced language model, to facilitate complex, long-context discussions. AI agents are configured with distinct roles and engage in structured dialogues to collaboratively evaluate and refine mitigation strategies. Analysis of 1,000 diverse agents revealed age as the most significant factor (importance: 0.14) influencing budget allocation, with younger participants (19–30 years) favoring immediate infrastructure investments while older participants (>61 years) preferred conservative strategies. The study demonstrates the potential of AI-driven simulations to replicate real-world collaborative environments, improving stakeholder engagement and enhancing the efficiency of hazard mitigation planning. The findings highlight the effectiveness of AI agents in multi-stakeholder decision-making processes, offering valuable insights for disaster risk reduction. This work contributes significantly to fostering more resilient, well-prepared communities through innovative approaches to decision-making.

UIOWA OA Agreement

AI-driven agents

budget optimization

decision-making

hazard mitigation

multi-agent system

water resources planning

Details

Title: Subtitle: AI-driven decision-making for water resource planning and hazard mitigation using automated multi-agents
Creators: Likith Kadiyala - University of Iowa, IIHR--Hydroscience and Engineering
Ramteja Sajja - University of Iowa
Yusuf Sermet - University of Iowa
Marian Muste - University of Iowa, IIHR--Hydroscience and Engineering
Ibrahim Demir - Tulane University
Resource Type: Journal article
Publication Details: Journal of hydroinformatics, Vol.27(7), pp.1217-1237
DOI: 10.2166/hydro.2025.042
ISSN: 1464-7141
eISSN: 1465-1734
Publisher: IWA Publishing; LONDON
Grant note: National Oceanic and Atmospheric Administration(NOAA): NA22NWS4320003
Thisproject was funded by the National Oceanic and Atmospheric Administration(NOAA) via a cooperative agreement with the University of Alabama (NA22NWS4320003) awarded to the Cooperative Institute for Research to Operations in Hydrology (CIROH).
Language: English
Date published: 06/23/2025
Academic Unit: Electrical and Computer Engineering; Civil and Environmental Engineering; IIHR--Hydroscience and Engineering; Injury Prevention Research Center; Geographical and Sustainability Sciences; Mechanical Engineering
Record Identifier: 9984843584902771

Metrics

22 Record Views

See more details