Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Intelligent 5S Audit: Application of Artificial Intelligence for Continuous Improvement in the Automotive Industry

Rafael da Silva Maciel, Lucio Veraldo

TL;DR

This work tackles the inefficiency and subjectivity of traditional 5S audits in automotive manufacturing by deploying a multimodal large language model–based automated audit system that analyzes images to assess Seiri, Seiton, Seiso, Seiketsu, and Shitsuke. The approach combines structured prompts, per-sense processing, and robust scoring to mirror human audits, achieving a Cohen's kappa of $\kappa = 0.75$ (substantial agreement) and delivering about a 50% reduction in audit time with ~99.8% cost savings. Key contributions include a complete end-to-end AI-aided auditing pipeline, an economic feasibility analysis with ROI trajectories, and implementation guidelines for Industry 4.0 environments. This research demonstrates a scalable, data-rich pathway for continuous improvement in automotive plants, enabling daily, objective, and traceable 5S assessments that support lean culture and rapid deviation correction.

Abstract

The evolution of the 5S methodology with the support of artificial intelligence techniques represents a significant opportunity to improve industrial organization audits in the automotive chain, making them more objective, efficient and aligned with Industry 4.0 standards. This work developed an automated 5S audit system based on large-scale language models (LLM), capable of assessing the five senses (Seiri, Seiton, Seiso, Seiketsu, Shitsuke) in a standardized way through intelligent image analysis. The system's reliability was validated using Cohen's concordance coefficient (kappa = 0.75), showing strong alignment between the automated assessments and the corresponding human audits. The results indicate that the proposed solution contributes significantly to continuous improvement in automotive manufacturing environments, speeding up the audit process by 50% of the traditional time and maintaining the consistency of the assessments, with a 99.8% reduction in operating costs compared to traditional manual audits. The methodology presented establishes a new paradigm for integrating lean systems with emerging AI technologies, offering scalability for implementation in automotive plants of different sizes.

Intelligent 5S Audit: Application of Artificial Intelligence for Continuous Improvement in the Automotive Industry

TL;DR

This work tackles the inefficiency and subjectivity of traditional 5S audits in automotive manufacturing by deploying a multimodal large language model–based automated audit system that analyzes images to assess Seiri, Seiton, Seiso, Seiketsu, and Shitsuke. The approach combines structured prompts, per-sense processing, and robust scoring to mirror human audits, achieving a Cohen's kappa of (substantial agreement) and delivering about a 50% reduction in audit time with ~99.8% cost savings. Key contributions include a complete end-to-end AI-aided auditing pipeline, an economic feasibility analysis with ROI trajectories, and implementation guidelines for Industry 4.0 environments. This research demonstrates a scalable, data-rich pathway for continuous improvement in automotive plants, enabling daily, objective, and traceable 5S assessments that support lean culture and rapid deviation correction.

Abstract

The evolution of the 5S methodology with the support of artificial intelligence techniques represents a significant opportunity to improve industrial organization audits in the automotive chain, making them more objective, efficient and aligned with Industry 4.0 standards. This work developed an automated 5S audit system based on large-scale language models (LLM), capable of assessing the five senses (Seiri, Seiton, Seiso, Seiketsu, Shitsuke) in a standardized way through intelligent image analysis. The system's reliability was validated using Cohen's concordance coefficient (kappa = 0.75), showing strong alignment between the automated assessments and the corresponding human audits. The results indicate that the proposed solution contributes significantly to continuous improvement in automotive manufacturing environments, speeding up the audit process by 50% of the traditional time and maintaining the consistency of the assessments, with a 99.8% reduction in operating costs compared to traditional manual audits. The methodology presented establishes a new paradigm for integrating lean systems with emerging AI technologies, offering scalability for implementation in automotive plants of different sizes.

Paper Structure

This paper contains 29 sections, 5 equations, 5 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Bibliographic Review
  3. 5S Methodology in Automotive Manufacturing
  4. Automation and Industry 4.0 in Automotive Manufacturing
  5. Artificial Intelligence Applied to Industrial Auditing Systems
  6. Methodologys
  7. Preparing the environment and collecting data
  8. Development of the Automated Audit System
  9. Prompt Engineering Architecture
  10. Specialized Processing by Sense
  11. Operational Robustness and Reliability
  12. Scoring and Ranking System
  13. Performance evaluation criteria
  14. Equations
  15. Analysis of operating costs
  16. ...and 14 more sections

Figures (5)

  • Figure 1: Architecture of the automated 5S audit system. It illustrates the flow from image input to report generation with built-in robustness mechanisms.
  • Figure 2: System performance by individual 5S sense. Seiri presented the highest agreement (83%), while Seiton represented the greatest analytical challenge (65%).
  • Figure 3: Comparison of audit duration: Automated = 20 min, Manual = 60 min.
  • Figure 4: Comparison of audit frequency: Automated approx. 100+, Manual = 20.
  • Figure 5: Comparison of audit cost: Automated = R$0.17, Manual = R$75.00.