The uniaxial compressive strength is the vital test used for designing rock engineering projects. Owing to the difficulties associated with preparing rock samples from weak, highly fractured and thinly bedded rocks, the predictive models were developed by many researchers. Recently researchers have begun evaluating the potential of neuro-fuzzy hybrid approach in several fields of studies. This paper presents the application of an adaptive neuro fuzzy inference system to predict uniaxial compressive strength of rocks using 126 data sets. For this purpose uniaxial compressive strength, Schmidt hammer, point load index, sound velocity, Physical properties, and Tensile Strength tests were applied. To control prediction performance of the obtained models, the root mean square error index was calculated as 13.65 from the neuro-fuzzy and 15.60 from the multiple regression model for test data set. The results are vastly promising. A comparative analysis proves that the adaptive neuro fuzzy inference system outperforms conventional methods.

This content is only available via PDF.
You can access this article if you purchase or spend a download.