Empirical Estimation of Uniaxial Compressive Strength of Rock: Database of Simple, Multiple, and Artificial Intelligence-Based Regressions

Empirical relationships for estimating Uniaxial Compressive Strength (UCS) of rock from other rock properties are numerous in literature. This is because the laboratory procedure for determination of UCS from compression tests is cumbersome, time consuming, and often considered expensive, especially for small to medium-sized mining engineering projects. However, these empirical models are scattered in literature, making it difficult to access a considerable number of them when there is need to select empirical model for estimation of UCS. This often leads to bias in estimated UCS data as there may be underestimation or overestimation of UCS, because of the site-specific nature of rock properties. Therefore, this study develops large database of empirical relationships between UCS and other rock properties that are reported in literatures. Statistical analysis was performed on the regression equations in the database developed. The typical ranges and mean of data used in developing the regressions, and the range and mean of their R-2 values were evaluated and summarised. Most of the regression equations were found to be developed from reasonable quantity of data with moderate to high R-2 values. The database can be easily assessed to select appropriate regression equation when there is need to estimate UCS for a specific site.

Automated summary

Numerous empirical relationships for estimating Uniaxial Compressive Strength (UCS) of rock from other rock properties are scattered in literature, making it challenging to select an appropriate model. This study focuses on developing a database of empirical relationships between UCS and other rock properties, analyzing regression equations statistically, and evaluating their consistency with reasonable data quantity and moderate to high R-2 values for accurate UCS estimation in specific sites.