Coupling Taguchi and Response Surface Methodologies for the Efficient Characterization of Jointed Rocks’ Mechanical Properties

Excerpt

Characterization of jointed rock mechanical properties has been a significant challenge in rock engineering due to its complex natural structure which consists of intact rock and discontinuities (Brady and Brown 2006). Some studies have explored the mechanical behaviour of jointed rocks from experimental viewpoint (Asadizadeh and Hossaini 2016; Asadizadeh et al. 2016; Asadizadeh and Rezaei 2019; Gehle and Kutter 2003; Ghazvinian et al. 2012, 2013; Lajtai 1969; Prudencio and Van-Sint-Jan 2007; Sarfarazi et al. 2014; Savilahti et al. 1990) and others used numerical modelling (Bahaaddini et al. 2013; Kim et al. 2007; Kulatilake et al. 1992; Sarfarazi et al. 2016; Vergara et al. 2016; Yang et al. 2016). Increase in the number of investigating parameters results in an exponential increase in the number of required experiments (Bahaaddini et al. 2016) while a decrease in the number of experiments can lead to a misunderstanding of the effects of independent parameters on the dependent parameters because of potential disregarding of accurate quantitative contribution of each independent parameter (Asadizadeh et al. 2018). Thus, whether laboratory investigation or numerical study is implemented, each needs a suitable design of experiments to ensure an optimum number of tests or modelling conditions is attained for statistically logical assessment of the mechanical properties of jointed rocks. Response Surface Methodology (RSM) (Box and Wilson 1951) and Taguchi technique (Taguchi and Konishi 1987) are two common approaches for the design of experiments that have been successfully utilized in rock engineering. …