Design of borehole deployments for slope stability analysis based on a probabilistic approach

Jing Sen Cai, Tian Chyi Jim Yeh, E. Chuan Yan, Rui Xuan Tang, Yong Hong Hao

Research output: Contribution to journalArticlepeer-review

6 Scopus citations


This study proposes a cross-correlation map-based borehole deployment approach for two-dimensional probabilistic slope stability analysis. This approach designs the layout of the proper number of boreholes based on the cross-correlation between the factor of safety and spatially variable soil strength every part of a slope. Numerically synthesized, undrained slopes are investigated as examples to illustrate the effectiveness of the proposed approach. Results demonstrate that the proposed approach is viable, and the cross-correlation maps are the appropriate metric for design slope borehole deployment. Using the cross-correlation maps, a small number of boreholes can sufficiently capture the large-scale heterogeneities that are critical to the slope stability. This information can help to identify the slip surface and improve the slope stability analysis. The small-scale heterogeneity, due to its short correlation structure or the residual covariance of the soil property field after conditioning using the borehole data, leads to a small amount of uncertainty in slope stability analysis. This small uncertainty could be vital to the slope stability analysis when the slope stability is close to the limit equilibrium state.

Original languageEnglish (US)
Article number103909
JournalComputers and Geotechnics
StatePublished - May 2021


  • Borehole deployment approach
  • Conditional analysis
  • Cross-correlation maps
  • Slope stability analysis
  • Spatial variability
  • Undrained shear strength

ASJC Scopus subject areas

  • Geotechnical Engineering and Engineering Geology
  • Computer Science Applications


Dive into the research topics of 'Design of borehole deployments for slope stability analysis based on a probabilistic approach'. Together they form a unique fingerprint.

Cite this