Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Smart Audit System Empowered by LLM

Xu Yao, Xiaoxu Wu, Xi Li, Huan Xu, Chenlei Li, Ping Huang, Si Li, Xiaoning Ma, Jiulong Shan

TL;DR

This work proposes a smart audit system empowered by large language models that streamlines audit procedures and optimizes resource allocation; a manufacturing compliance copilot that enhances data processing, retrieval, and evaluation for a self-evolving manufacturing knowledge base; and a Re-act framework commonality analysis agent that provides real-time, customized analysis to empower engineers with insights for supplier improvement.

Abstract

Manufacturing quality audits are pivotal for ensuring high product standards in mass production environments. Traditional auditing processes, however, are labor-intensive and reliant on human expertise, posing challenges in maintaining transparency, accountability, and continuous improvement across complex global supply chains. To address these challenges, we propose a smart audit system empowered by large language models (LLMs). Our approach introduces three innovations: a dynamic risk assessment model that streamlines audit procedures and optimizes resource allocation; a manufacturing compliance copilot that enhances data processing, retrieval, and evaluation for a self-evolving manufacturing knowledge base; and a Re-act framework commonality analysis agent that provides real-time, customized analysis to empower engineers with insights for supplier improvement. These enhancements elevate audit efficiency and effectiveness, with testing scenarios demonstrating an improvement of over 24%.

Smart Audit System Empowered by LLM

TL;DR

This work proposes a smart audit system empowered by large language models that streamlines audit procedures and optimizes resource allocation; a manufacturing compliance copilot that enhances data processing, retrieval, and evaluation for a self-evolving manufacturing knowledge base; and a Re-act framework commonality analysis agent that provides real-time, customized analysis to empower engineers with insights for supplier improvement.

Abstract

Manufacturing quality audits are pivotal for ensuring high product standards in mass production environments. Traditional auditing processes, however, are labor-intensive and reliant on human expertise, posing challenges in maintaining transparency, accountability, and continuous improvement across complex global supply chains. To address these challenges, we propose a smart audit system empowered by large language models (LLMs). Our approach introduces three innovations: a dynamic risk assessment model that streamlines audit procedures and optimizes resource allocation; a manufacturing compliance copilot that enhances data processing, retrieval, and evaluation for a self-evolving manufacturing knowledge base; and a Re-act framework commonality analysis agent that provides real-time, customized analysis to empower engineers with insights for supplier improvement. These enhancements elevate audit efficiency and effectiveness, with testing scenarios demonstrating an improvement of over 24%.

Paper Structure

This paper contains 19 sections, 6 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Works
  3. Digitization in Quality Management Systems (QMS) and Supplier Management
  4. Domain-Specified Retrieval-Augmented Generation (RAG)
  5. Re-Act Agent
  6. Methodologies
  7. Overview
  8. Dynamic Risk Assessment in Data Collection
  9. Manufacturing Compliance Copilot in Knowledge Base Consolidation
  10. Commonality Analysis Agent in Data Analysis
  11. Experiments
  12. Setup
  13. Results
  14. Use Cases Demonstrations
  15. Smart audit app
  16. ...and 4 more sections

Figures (6)

  • Figure 1: Smart Audit System Framework a) An initial phase utilizing natural language processing to optimize data collection through dynamic risk assessment; b) The integration of a compliance copilot system into existing manufacturing knowledge bases, employing a multi-task, instruction-guided analysis to convert raw data into actionable insights; c) Deployment of a Commonality Study Agent that enhances real-time accountability and process improvements using a variant consistency approach.
  • Figure 2: Dynamic Risk Assessment in Data Collection Pipeline
  • Figure 3: Manufacturing Compliance Copilot Structure
  • Figure 4: Commonality Analysis Agent Data Pipeline
  • Figure 5: Smart Audit App: Facilitating Data Collection with an Embedded Dynamic Risk Assessment Model
  • ...and 1 more figures