Integration of YOLO (You Only Look Once) AI and Lean methodology for safer and more efficient work environments. A Systematic Review

Abstract

The integration of computer vision technologies with continuous improvement methodologies has emerged as a key strategy to enhance safety and efficiency in industrial environments. This study evaluates the combined impact of the YOLO algorithm and Lean methodology on PPE supervision and operational performance in the manufacturing sector, using a Systematic Literature Review (SLR) guided by the PICO and PRISMA frameworks. A total of 69 relevant studies (2020–2025) were analyzed from databases including Scopus, IEEE Xplore, and Web of Science. Findings reveal that YOLO, particularly in its recent versions like YOLOv7, achieves over 95% accuracy in real-time PPE detection, outperforming traditional methods such as RFID sensors and classic vision algorithms. Meanwhile, Lean tools such as 5S+1, Kaizen, and Poka-Yoke prove effective in error reduction, process standardization, and fostering a preventative culture. The synergy between YOLO and Lean enables precise automated supervision, improves regulatory compliance traceability, and lowers exposure to occupational hazards. This combined approach presents an innovative and efficient solution to reinforce workplace safety and operational performance. However, limitations persist due to the lack of empirical integration studies and technological adoption challenges faced by small enterprises, mainly driven by economic and technical constraints. These gaps highlight opportunities for future research focused on scalable and sustainable implementation.