Reverse k-ratio and geological complexities have significantly contributed to design challenges and instability within the rockmass. The rock engineering design process within this shaft had to consider that mining had to occur in a rockmass that included large blocks of ground truncated by major geological weaknesses, reef planes with different strike, dip orientations, and a rockmass impaled with joint and fracture planes. The off-reef tunnels in the weaker shale rock type tend to squeeze and are supported with ring sets when the deformation occurs within these tunnels. Designing tunnels near this rock mass considered the presence of a weak, highly altered Westonaria Formation Lava, which directly over-lays the high-grade Ventersdorp Contact Reef conglomerate within the shaft. Westonaria Formation Lava has significantly contributed to instability, posing an immediate risk to shaft barrel stability and a medium- to long-term risk to continued ore extraction within the excavating environment.

The stability graph method is a widely used empirical method for dimensioning open rooms/stopes and support design based on stope geometry and stability number. Despite numerous developments, such as expanding databases and introducing new concepts and factors, a comprehensive understanding of its underlying mechanisms and universal applicability necessitates thorough numerical and theoretic analyses across various scenarios. This paper presents a rigorous investigation employing numerical models featuring multiple joint sets to replicate the structurally controlled failure of open stopes, a predominant failure type encountered in the stability graph databases. Detached blocks within these models are treated as overbreak and identified by defining thresholds for normal and shear displacements on joint planes. The concept of equivalent linear overbreak/slough (ELOS) is referred to in the numerical models to quantify the failure, similar to the ELOS stability graph method. The results obtained from numerical models and empirical method for the cases with different stope dips and sizes, and different critical joint set orientations have been compared to evaluate the performance of stability graph under different scenarios. It is found that the agreements between the numerical and empirical results on different surfaces of the open stopes are different. Notably, the adaptation of a modified stress factor A significantly enhances agreement, particularly for the hanging wall, characterized by a lower confinement stress state. Furthermore, larger ELOS values on the back surface of stopes with a critical joint set angle of 45 degrees are identified compared to those with a critical joint set angle of 30 degrees. It has opposite trend to the stability graph results. To address this inconsistency, modified factor B values are proposed for different surfaces based on normalized analyses of numerical models with varying critical joint set angles. In conclusion, the numerical simulation results align well with stability graph outcomes when using modified factors, A and B. This research has improved our understanding of the empirical stability graph method and promotes its reliability in predicting the stability state and unplanned dilution in open stopes. These insights are significant for mining engineers and practitioners seeking more accurate and robust stope stability assessments in their operations.

In opencast mines, overburden removal is the first step for mineral extraction, which is disposed of as dump slopes made of several benches. Shear strength is the most significant factor for establishing the slope stability of the dump. The laboratory techniques to determine the shear parameters, such as cohesion (C) and internal friction angle (Φ), don’t represent the actual behavior of the samples. In the present work, a large-scale in-situ direct shear apparatus is fabricated for determining the shear parameters of a dump in constant normal load (CNL) condition. The apparatus was tested over dumps of iron-ore mines in Odisha, India. The peak shear stresses corresponding to various vertical stresses, viz. 25, 50, 75, and 150 kPa, were found to determine a Mohr-Coulomb failure envelope. The in-situ C and Φ were measured to be 31.41 kPa and 48.99⁰, respectively. These in-situ tests give realistic data, which is very helpful for the optimistic design of overburden dumps.

The limit equilibrium model in boundary element codes has become a popular method to simulate the behaviour and failure of pillars in underground workings. Albeit good results have been obtained through this model, the calibration of this model is cumbersome due to the multitude of parameters that require calibration. An alternative solution was presented to verify and calibrate the model through the use of physical modelling in a laboratory. For the experiments, an artificial pillar material was used and cubes were poured using the standard 100x100mm civil engineering moulds. The friction angle between the artificial “pillar” and the platens of the testing machine was varied by using soap and sandpaper. Different modes of failure were observed depending on the friction angle. The results of the preliminary numerical modelling indicated that the model is able to simulate the stress-strain behaviour of the laboratory models, thereby verifying that the limit equilibrium model appears to be a useful approximation of the pillar failure. This paper further investigates the numerical modelling of the laboratory experiments conducted.

The complex geometric morphology of single rough-walled rock fractures, coupled with the occurrence of nonlinear flow, adds complexity to the fracture flow process. Despite decades of research on nonlinear flow behavior in single rock fractures, existing models still fall short of adequately capturing such behavior during shearing. In this study, a series of coupled shear-flow tests are conducted on single rock fractures under constant normal loads. The results show that the Forchheimer equation effectively describes nonlinear flow, with its nonlinear coefficients associated with fracture geometries. The evolution of fracture geometries induced by shearing is quantified and its impact on nonlinear flow is considered. An empirical equation is then proposed, incorporating the peak asperity height and hydraulic aperture, to evaluate the Forchheimer nonlinear coefficient. The proposed equation is validated through experimental results, demonstrating its effectiveness in characterizing nonlinear flow behavior in rock fractures during shearing.

Rock mass is the host media in a wide range of geotechnical applications. The failure behaviour of rock mass is complex and strongly influenced by various geological structures, from micro to macro scale. To have a proper structural design and safe construction, a thorough understanding of failure mechanism around an underground excavation is essential. That includes an investigation of the cracking processes of rock mass and, in particular, how micro-cracking around the excavation progresses to macro-cracking of rock mass. In order to reveal the rock mass cracking process of an underground excavation, laboratory investigations on real rocks containing an opening, can provide a proper understanding of its complex behavior. In this regard, different advanced techniques can be implemented to monitor the cracking process around the excavation in real time and to explore the rock mass behavior more deeply. This paper aims to experimentally investigate the fracture evolution and damage behavior of rock specimens containing an opening under the uniaxial compression, incorporating digital image correlation (DIC) method and acoustic emission (AE) technique. For this purpose, uniaxial compression tests were conducted on granitic rock blocks containing a circular opening. In addition, four strain gauges were installed in as many directions around the opening to record the local strains. The synchronized AE output parameters and DIC plots with the stress and displacement data extracted from the servo-controlled loading frame, together with strain gauge data have been compared and analyzed in detail to reveal the mechanisms of crack coalescence. Finally, different criteria have been implemented to estimate damage stress threshold values. Combined analysis of AE parameters, DIC and stress-strain data allows to identify and discriminate different cracking stages including crack initiation, growth, coalescence and damage in tested specimens and their results are mutually consistent. AE results provide with reliable information to characterize the fracture evolution stages and different stress thresholds. Likewise, the local strain levels recorded by strain gauges are in a fair good agreement with the strain field measured with DIC. The type of crack initiation and failure around circular opening are also well characterized and in accordance to the theoretical crack distribution around excavations. The results of this study provide significant insights of the cracking processes in rock mass around an opening.