A Basic Study on the Prediction of Collapse of Tunnels Using Artificial Neural Network

인공신경망 기법을 이용한 터널 붕괴 예측에 관한 기초 연구

Kim, Hong-Heum;Lim, Heui-Dae

  • Received : 2015.10.28
  • Accepted : 2016.01.30
  • Published : 2016.02.29


Collapse of a tunnel can occur anytime, anywhere due to the special characteristics of tunnel structures and unexpected geological conditions during construction. Tunnel collapse will lead to economic losses and casualties. So various studies are continually being conducted to prevent economic losses, casualties and accidents. In this study, we analyzed data from 56 domestic construction tunnel collapse sites, and input factors to be applied to the artificial neural network were selected by the sensitivity analysis. And for the artificial neural network model design studies were carried out with the selected input factors and optimized ANN model to predict the type of tunnel collapse was determined. By using it, in 12 sites where tunnel collapse occurred applicability evaluation was conducted. Thus, the tunnel collapse type predictability was verified. These results will be able to be used as basic data for preventing and reinforcing collapse in the tunnel construction site.


Collapse;Tunnel;ANN (Artificial neural network);Sensitivity analysis;Training


  1. Anderson, J. (1997), "HSE Report on NATM Safety", Tunnel & Tunnelling, March, pp.49-51.
  2. Caudill, M. (1988), "Neural Networks Primer, Part III", AI EXPERT, 3(6), pp.53-59.
  3. Hwang, M.G. (2004), "Application of Artificial Neural Networks to Soil Flushing Method", Master's thesis, University of Myongji, pp.1-24.
  4. Jeong, Y.S. (2008), "Development of Tunnel Collapse Risk Index by Neural Network", Master's thesis, University of Hoseo, pp. 20-40.
  5. Kim, J.K. (2011), "A Case Study on the Estimation of a Proper Tunnel Support Pattern Using Back Analysis based on Artificial Neural Network, Doctoral thesis, University of Suwon, pp.20-27.
  6. Kim, Y.Y. (2011), "A Study on Quantified Management for Tunnel Construction Hazard based on KICT Tunnel Hazard Index (KTHIndex)", Master's thesis, University of Dongguk, pp.10-12.
  7. Korean Institutue of Construction Technology (2009), "Development of Technologies for Minimizing and Prevention the Disaster on Tunnel Construction VI", report, KICT 2009-071, pp.43-64.
  8. Korean Tunnelling Association (2010), "Case Histories of Tunnel Collapse", CIR, pp.93-263.
  9. Lee, S.B. and Im, Y.D. (2002), "Fuzzy Neural Network Genetic Evolution", INSOL MEDIA, pp.107-144.
  10. Lee, S.G., Kim, N.Y., and Jeon B.H. (2007), "A Study on the Collapse Pattern of Road Tunnel under Construction", The Journal of Engineering Geology, Vol.17, No.1, March, pp.115-123.
  11. Nam, K.M. (2008), "A Study on the Estimation of the Blast Vibration Velocity in the Structure Using the Artificial Neural Network", Master's Thesis, Chungnam National University, pp.21-52.
  12. Shin, H.S., Kwon, Y.C., Jung, Y.S, Bae, G.J., and Kim, Y.Y. (2009), "Methodology for Quantitative Hazard Assessment for Tunnel Collapses based on Case Histories in Korea", International Journal of Rock Mechanics and Mining Science, Vol.45, No.5, pp.1072-1087.

Cited by

  1. Prediction of Asphalt Pavement Service Life using Deep Learning vol.20, pp.2, 2018,