ABSTRACT
In this study, a fitting method for the Hoek-Brown parameters is presented, where the measurement of error is based on the distance between the yield surface and the experimental points in the stress vector space. The distance between experimental data and the 3d curve is defined as the minimum distance obtained through closest point projection. A least squares method based on this distance is used with a grid search algorithm. The algorithm is tested on various data sets that can be found in the literature. The results are compared with the values obtained by other authors, and both a standard Hoek Brown failure criterion and a modified smooth version are fitted.
