Safety and Health at Work

Occupational Safety and Health Research Institute, Korea Occupational Safety and Health Agency (산업안전보건연구원)
권호 표지
DOI QR코드

DOI QR Code

Management Architecture With Multi-modal Ensemble AI Models for Worker Safety

  • Dongyeop Lee (Team of Occupational Safety, Convergence Technology Lab, KEPCO Research Institute) ;
  • Daesik, Lim (Team of Occupational Safety, Convergence Technology Lab, KEPCO Research Institute) ;
  • Jongseok Park (Team of Occupational Safety, Convergence Technology Lab, KEPCO Research Institute) ;
  • Soojeong Woo (Team of Occupational Safety, Convergence Technology Lab, KEPCO Research Institute) ;
  • Youngho Moon (Team of Occupational Safety, Convergence Technology Lab, KEPCO Research Institute) ;
  • Aesol Jung (Team of Occupational Safety, Convergence Technology Lab, KEPCO Research Institute)
  • Received : 2024.04.04
  • Accepted : 2024.04.30
  • Published : 2024.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Introduction: Following the Republic of Korea electric power industry site-specific safety management system, this paper proposes a novel safety autonomous platform (SAP) architecture that can automatically and precisely manage on-site safety through ensemble artificial intelligence (AI) models. The ensemble AI model was generated from video information and worker's biometric information as learning data and the estimation results of this model are based on standard operating procedures of the workplace and safety rules. Methods: The ensemble AI model is designed and implemented by the Hadoop ecosystem with Kafka/NiFi, Spark/Hive, HUE, and ELK (Elasticsearch, Logstash, Kibana). Results: The functional evaluation shows that the main function of this SAP architecture was operated successfully. Discussion: The proposed model is confirmed to work well with safety mobility gateways to provide some safety applications.

Keywords

Acknowledgement

We would like to thank the construction site workers at KEPCO offices for their cooperation in the field verification of the research results.

References

  1. Floyd II HL. Reflections on a 50-year journey in electrical safety: further steps toward risk reduction. IEEE Ind Appl Mag 2023;29:59-65. 
  2. Sun X, He Y, Wu D, Huang JZ. Survey of distributed computing frameworks for supporting big data analysis. Big Data Min Anal 2023;6:154-69. 
  3. Asaithambi SPR, Venkatraman R, Venkatraman S. A thematic travel recommendation system using an augmented big data analytical model. Tech 2023;11:1-28. 
  4. Calderon G, del Campo G, Saavedra E, Santamaria A. Monitoring framework for the performance evaluation of an IoT platform with Elasticsearch and Apache Kafka. Inf Syst Front 2023;4:1-17. 
  5. Hussain M. YOLO-v5 variant selection algorithm coupled with representative augmentations for modelling production-based variance in automated lightweight pallet racking inspection. Big Data Cogn Comput 2023;7:1-15. 
  6. Lee J, Cho S, Yu D. Comparative study of time series anomaly detection based on generative deep learning models. JKDISS 2023;34:377-94 [in Korean].