A Ranking Distance Analysis for Performance Assessment of UCS Versus SPT-N Correlations

Abstract

It is important to know the reliability of the correlations unconfined compressive strength (UCS) versus blow count of standard penetration test (SPT-N) available in the literature for foundation design of constructions. However, selection of best performed equation verified with local data always becomes a primary issue for the site of interest foundation constructed. Thus, this paper investigates the performances of some UCS versus SPT-N correlations for the applicability to fine-grained soils in an interested area. A novel ranking measure called ranking distance (RD) that accounts equal weighting to the accuracy and precision of estimation has been used for the performance assessment of the equations, in comparison with alternative measures. The RD index is a comprehensive ranking measure for performing the dependability of an empirical equation with regard to the accuracy and precision, using the ratios of estimated and measured UCS. Dependent upon the analysis results in the study, the performances of equations have been ranked high, moderate, low and unacceptable (the worst equations). The equations have also been categorized whether they influence more weighting to precise or accuracy, or equal accuracy and precise during the estimations of UCS. From the results, it is concluded that the best performed estimation based on the RD is proposed by the equation of low plastic clay (Sivrikaya and Togrol, in: 5th international congress on advances in civil engineering, Istanbul, Turkey, pp 943–952, 2002; Eng Geol 86:52–69, 2006). Consequently, the study has successfully demonstrated the use of RD for assessing the best equation by confirming its results with the alternative methodologies compared. This could be beneficial for the experimenters in practice to attempt the validations of different empirical equations with different parameter through a good ranking procedure.