DOI QR코드

DOI QR Code

Real-time Fall Accident Prediction using Random Forest in IoT Environment

사물인터넷 환경에서 랜덤포레스트를 이용한 실시간 낙상 사고 예측

  • Chan-Woo Bang (Department of Information Communication Enginerring, Seowon University) ;
  • Bong-Hyun Kim (Department of Computer Engineering, Seowon University)
  • 방찬우 (서원대학교 정보통신공학과) ;
  • 김봉현 (서원대학교 컴퓨터공학과)
  • Received : 2024.07.03
  • Accepted : 2024.08.14
  • Published : 2024.08.31

Abstract

As of 2023, the number of accident victims in the domestic construction industry is 26,829, ranking second only to other businesses (service industries). The accident types of casualties in all industries were falls (29,229 people), followed by falls (14,357 people). Based on the above data, this study attaches sensors to hard hats and insoles to predict fall accidents that frequently occur at construction sites, and proposes smart safety equipment that applies a random forest algorithm based on the data collected through this. The random forest model can determine fall accidents in real time with high accuracy by generating multiple decision trees and combining the predictions of each tree. This model classifies whether a worker has had a fall accident and the type of behavior through data collected from the MPU-6050 sensor attached to the hard hat. Fall accidents that are primarily determined from hard hats are secondarily predicted through sensors attached to the insole, thereby increasing prediction accuracy. It is expected that this will enable rapid response in the event of an accident, thereby reducing worker deaths and accidents.

2023년 기준 국내 건설업에서 발생한 사고 재해자 수는 26,829명으로 기타의 사업(서비스업)에 이어 두 번째에 해당한다. 전 업종 재해자 사고 유형으로는 넘어짐(29,229명), 떨어짐(14,357명) 순으로 이루어져 있다. 위 자료를 토대로 본 연구에서는 건설 현장에서 빈번하게 발생하는 낙상 사고를 예측하기 위해 안전모와 깔창에 센서를 부착하고, 이를 통해 수집된 데이터를 바탕으로 랜덤 포레스트 알고리즘을 적용한 스마트 안전 장비를 제안한다. 랜덤 포레스트 모델은 여러 결정 트리를 생성하여 각 트리의 예측을 종합함으로써 높은 정확도로 낙상 사고를 실시간으로 판별할 수 있다. 이 모델은 안전모에 부착된 MPU-6050 센서에서 수집된 데이터를 통해 노동자의 낙상 사고 여부와 행동 유형을 분류한다. 안전모로부터 일차적으로 판별된 낙상사고는 깔창에 부착된 센서를 통해 이차적으로 예측하여, 예측 정확도를 높인다. 이를 통해 사고 발생 시 신속한 대응이 가능하여 노동자의 사망 및 재해사고를 줄일 수 있다고 기대한다.

Keywords

References

  1. H.H. Sim and K.S. Kang, "A study on the death accident analysis of ladder and prevention measures for fall accidents," Journal of the Korea Safety Management & Science, Vol.19, No.4, pp.95-104, 2017.
  2. C. W. Bang and B. H. Kim, "Design of smart safety equipment platform for construction site workers," Journal of Next-generation Convergence Technology Association, Vol.6, No.7, pp.1182-1188, 2022. https://doi.org/10.33097/JNCTA.2022.06.07.1182
  3. Y. Seo, S. Kim, D. Lee, J. Jung, and D. Kim, "Correlation analysis between awareness of the serious accidents and safety consciousness of construction workers under the act on the punishment of serious accidents," Korean Journal of Construction Engineering and Management, Vol.25, No.3, pp.47-57, 2024. https://doi.org/10.6106/KJCEM.2024.25.3.047
  4. K. W. Lee, Y. J. Won, Y. B. Song, and K. S. Cho, "Development of machine learning ensemble model using artificial intelligence," Journal of the Korean Society for Heat Treatment, Vol.34, No.5, pp.211-217, 2021. https://doi.org/10.12656/JKSHT.2021.34.5.211
  5. B.U. Jeon, J.S. Kang, and K. Chung, "AutoML and CNN-based soft-voting ensemble classification model for road traffic emerging risk detection," Journal of Convergence for Information Technology, Vol.11, No.7, pp.14-20, 2021. https://doi.org/10.22156/CS4SMB.2021.11.07.014
  6. S. Park, "Malicious insider detection using boosting ensemble methods," Journal of the Korea Institute of Information Security & Cryptology, Vol.32, No.2, pp.267-277, 2022.
  7. S.H. Min, "Bankruptcy prediction using an improved bagging ensemble," Journal of Intelligence and Information Systems, Korea Intelligent Information System Society, December 30, 2014.
  8. H. I. Kim, Y. S. Lee, and B. Kim, "Real-time flood prediction applying random forest regression model in urban areas," Journal of Korea Water Resources Association, Vol.54, Spc1, pp.1119-1130, 2021.
  9. J. Hong and S.J. Jeon, "Prediction of safety grade of bridges using the classification models of decision tree and random forest," KSCE Journal of Civil and Environmental Engineering Research, Vol.43, No.3, pp.397-411, 2023. https://doi.org/10.12652/KSCE.2023.43.3.0397
  10. J. A. Chun, H.J. Lee, S.H. Im, D. Kim, and S.S. Baek, "Comparative assessment of frost event prediction models using logistic regression, random forest, and LSTM networks," Journal of Korea Water Resources Association, Vol.54, No.9, pp.667-680, 2021. https://doi.org/10.3741/JKWRA.2021.54.9.667
  11. D.H. Hwang and Y.S. Seo, "A study on accounting fraud detection using neural network and random forest," Proceedings of the Korea Information Processing Society Conference, pp.692-693, 2023.
  12. I. Mahmoud, I. Saidi, and C. Bouzazi, "Design of an IOT system based on face recognition technology using ESP32-CAM," International Journal of Computer Science and Network Security, Vol.22, No.8, pp.1-6, 2022. https://doi.org/10.22937/IJCSNS.2022.22.8.1
  13. Y.H. Kim, "A study of vibration energy harvesting for the bimorph piezoelectric sensor," Journal of Sensor Science and Technology, Vol.19, No.4, pp.313-319, 2010. https://doi.org/10.5369/JSST.2010.19.4.313
  14. G.C. Park, S.H. Kim, S.-W. Baik, J.-H. Kim, and G.-R. Jeon, "Discrimination of fall and fall-like ADL using tri-axial accelerometer and bi-axial gyroscope," Journal of Sensor Science and Technology, Vol.26, No.1, pp.7-14, 2017. https://doi.org/10.5369/JSST.2017.26.1.7
  15. Y. Gao and N.H. Kim, "Direction information concerned algorithm for removing Gaussian noise in images," Journal of Information and Communication Convergence Engineering, The Korean Institute of Information and Communication Sciences, December 31, 2011.