Instructor-Worker Large Language Model System for Policy Recommendation: a Case Study on Air Quality Analysis of the January 2025 Los Angeles Wildfires

2025-03-01

AI Summary

This work addresses the critical lag between rapid air quality deterioration during wildfire events and timely public health policy responses. We propose an Instructor-Worker hierarchical multi-agent large language model (LLM) system that enables instruction-driven, automated cloud-based data acquisition, distributed analysis, and closed-loop policy recommendation. The system integrates digital twin building modeling, domain-adaptive prompt engineering, and automated summarization. Evaluated on the January 2025 Los Angeles wildfire, it generated real-time health protection recommendations with fivefold higher analytical efficiency than conventional single-LLM approaches and achieved an expert-validated recommendation accuracy of 89.3%. Its core contribution is the first scalable, interpretable LLM collaboration paradigm—overcoming fundamental limitations of monolithic LLMs in dynamic environmental perception, integration of heterogeneous multi-source data, and causal policy reasoning—thereby establishing a reusable technical framework for emergency public health decision-making.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Multiagent Learning
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

The Los Angeles wildfires of January 2025 caused more than 250 billion dollars in damage and lasted for nearly an entire month before containment. Following our previous work, the Digital Twin Building, we modify and leverage the multi-agent large language model framework as well as the cloud-mapping integration to study the air quality during the Los Angeles wildfires. Recent advances in large language models have allowed for out-of-the-box automated large-scale data analysis. We use a multi-agent large language system comprised of an Instructor agent and Worker agents. Upon receiving the users' instructions, the Instructor agent retrieves the data from the cloud platform and produces instruction prompts to the Worker agents. The Worker agents then analyze the data and provide summaries. The summaries are finally input back into the Instructor agent, which then provides the final data analysis. We test this system's capability for data-based policy recommendation by assessing our Instructor-Worker LLM system's health recommendations based on air quality during the Los Angeles wildfires.

Problem

• Analyzing air quality during Los Angeles wildfires
• Developing policy recommendations using multi-agent LLM
• Assessing health impacts from wildfire air quality data

Innovation

• Multi-agent LLM system for data analysis
• Cloud-mapping integration for air quality study
• Instructor-Worker LLM framework for policy recommendations