Root Cause Analysis in Food Safety Incidents with Large Language Models
Ensuring food safety is a critical global challenge, as foodborne illnesses and contamination incidents continue to pose risks to public health, supply chains, and regulatory compliance. The ability to accurately identify the root causes of contamination, assess risk factors, and implement effective mitigation strategies is crucial for preventing outbreaks and minimizing their impact. However, the vast amount of available food safety data—including scientific research, regulatory reports, and historical incident records—makes manual analysis time-consuming and inefficient.
This use case explores the application of Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) in conducting root cause analysis for food safety incidents. Specifically, we leverage two state-of-the-art LLMs—Qwen2.5-1.5B-Instruct and DeepSeek-R1-Distill-Qwen-1.5B—to process and analyze past food safety cases (2022–2023) and generate insights based on structured data retrieval. By integrating AI-powered models with a comprehensive food safety intelligence database, we aim to improve the efficiency, accuracy, and scalability of contamination investigations.