Probabilistic modelling of joint orientation

Pinnaduwa H.S.W. Kulatilake, Tien H. Wu, Deepa N. Wathugala

Research output: Contribution to journalArticle

25 Scopus citations

Abstract

Observed frequencies of joint orientations are subject to error due to sampling bias. This error should be corrected before statistical inference is made on the distribution of orientation. Corrections (weighting functions) are developed for sampling bias in orientation for finite joints of different sizes and shapes intersecting rectangular exposures. Chi‐square goodness‐of‐fit procedures available for hemispherical normal and bivariate normal distributions are modified to make them applicable for both raw data and data corrected for sampling bias. The aforementioned corrections and procedures were applied to a joint orientation cluster to study the effect of (a) joint orientation, (b) joint size and (c) joint shape, on the statistical distribution of the orientation. The influences of all these aforementioned factors were found to be significant. However, at present, joint sizes and shapes are not measured in field joint surveys. Therefore, it is suggested to make attempts to obtain joint sizes and shapes in field joint mapping surveys. Since the currently available probability distributions are not adequate to represent all joint orientation distributions, it is suggested either to look for new probability distributions or to develop procedures to use empirical distributions in modelling orientation distributions.

Original languageEnglish (US)
Pages (from-to)325-350
Number of pages26
JournalInternational Journal for Numerical and Analytical Methods in Geomechanics
Volume14
Issue number5
DOIs
StatePublished - Jul 1990

ASJC Scopus subject areas

  • Computational Mechanics
  • Materials Science(all)
  • Geotechnical Engineering and Engineering Geology
  • Mechanics of Materials

Fingerprint Dive into the research topics of 'Probabilistic modelling of joint orientation'. Together they form a unique fingerprint.

  • Cite this