SOPRAG: Multi-view Graph Experts Retrieval for Industrial Standard Operating Procedures

TL;DR

SOPRAG is a specialized multi-view graph expert retrieval framework designed to improve industrial SOP retrieval by addressing structural, logical, and domain-specific challenges, leading to significant performance gains.

Contribution

It introduces a novel MoE-based framework with specialized graph experts, a Procedure Card layer, and an LLM-guided gating mechanism for effective SOP retrieval in industrial settings.

Findings

Outperforms existing RAG baselines in accuracy and utility

Achieves perfect execution scores in real-world tasks

Demonstrates robustness across four industrial domains

Abstract

Standard Operating Procedures (SOPs) are essential for ensuring operational safety and consistency in industrial environments. However, retrieving and following these procedures presents unique challenges, such as rigid proprietary structures, condition-dependent relevance, and actionable execution requirement, which standard semantic-driven Retrieval-Augmented Generation (RAG) paradigms fail to address. Inspired by the Mixture-of-Experts (MoE) paradigm, we propose SOPRAG, a novel framework specifically designed to address the above pain points in SOP retrieval. SOPRAG replaces flat chunking with specialized Entity, Causal, and Flow graph experts to resolve industrial structural and logical complexities. To optimize and coordinate these experts, we propose a Procedure Card layer that prunes the search space to eliminate computational noise, and an LLM-Guided gating mechanism that dynamically weights these experts to align retrieval with operator intent. To address the scarcity of domain-specific data, we also introduce an automated, multi-agent workflow for benchmark construction. Extensive experiments across four industrial domains demonstrate that SOPRAG significantly outperforms strong lexical, dense, and graph-based RAG baselines in both retrieval accuracy and response utility, achieving perfect execution scores in real-world critical tasks.