KRAIL: A Knowledge-Driven Framework for Base Human Reliability Analysis Integrating IDHEAS and Large Language Models

TL;DR

KRAIL is a novel framework that combines IDHEAS, large language models, and knowledge graphs to improve the efficiency and accuracy of human error probability estimation in reliability analysis.

Contribution

The paper introduces a semi-automated, knowledge-driven approach integrating LLMs and knowledge graphs for base HEP estimation, reducing reliance on subjective expert judgment.

Findings

Demonstrates superior performance in HEP estimation under partial information

Uses knowledge graphs for efficient data retrieval and processing

Validates effectiveness on authoritative human reliability datasets

Abstract

Human reliability analysis (HRA) is crucial for evaluating and improving the safety of complex systems. Recent efforts have focused on estimating human error probability (HEP), but existing methods often rely heavily on expert knowledge,which can be subjective and time-consuming. Inspired by the success of large language models (LLMs) in natural language processing, this paper introduces a novel two-stage framework for knowledge-driven reliability analysis, integrating IDHEAS and LLMs (KRAIL). This innovative framework enables the semi-automated computation of base HEP values. Additionally, knowledge graphs are utilized as a form of retrieval-augmented generation (RAG) for enhancing the framework' s capability to retrieve and process relevant data efficiently. Experiments are systematically conducted and evaluated on authoritative datasets of human reliability. The experimental results of the proposed methodology demonstrate its superior performance on base HEP estimation under partial information for reliability assessment.