Uncertainty-Based Analysis for Agreement of Tensile-Strength Measurement Procedures

TÜTMEZ B., Unver B.

GEOTECHNICAL TESTING JOURNAL, vol.40, no.3, 2017 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 40 Issue: 3
  • Publication Date: 2017
  • Doi Number: 10.1520/gtj20160217
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Keywords: tensile strength, measurement uncertainty, standard averaging, maximum likelihood criterion
  • Hacettepe University Affiliated: Yes


Rocks fail because of compression, shear, tensile and/or a combination of these stresses. Rocks are more vulnerable to tensile stresses as they can stably accommodate only a fraction of compressive stresses in tensional loading. Therefore, reliable determination of tensile strength of rocks is important. Tensile strength of rocks that can be determined by both direct and indirect tensile testing methods have great importance in rock mechanics. Although these methods have different procedures and quite a change in results, an agreement between the methods is necessary. In addition, properly appraising the data provided by direct and indirect measurements has critical importance for applying the knowledge thereby acquired. From this point, based on the conventional measurement uncertainty framework, two weighting methods are suggested to combine the tensile strength methods. Thus, an uncertainty-based agreement between the direct and indirect tensile-strength measurements has been made. The results showed that the weighted uncertainties of the proposed methods are lower than the best single measurement uncertainties. In addition, the weighted averages and corresponding uncertainties provide additional information and unique outcomes for the users. A reliable correlation between indirect and direct tensile-strength results is certainly useful for assessing tensile strength of rocks.