A risk based methodology to improve the definition of geotechnical design sectors in slope design

Abstract

In open pit mines, the slope design process is widely utilised in a relatively standard format. Data from geological, structural, rock mass and hydrogeological models are used to formulate and populate geotechnical models. The volume of interest is divided into geotechnical domains, wherein structural and rock strength data are key inputs to define conceptual failure modes or models. Subdivision of domains, which contain structurally-controlled and/or anisotropic materials, into design sectors should result in practical slope designs. However, in the planning of open pit mines, it is often difficult to determine the shapes and sizes of slope design sectors prior to their analysis, particularly in structurally-complex geological environments. The most challenging aspects of design, in moderately- to heavily-jointed rock masses, relate to the impacts of in-situ geological features on rock mass behaviour. In addition, joint density and persistence must be considered relative to the volume of interest, as rock mass properties can exhibit significant scale-dependence. This problem is exacerbated in sectors encompassing significant strike changes of lithological units or contacts. This variability in strike is often poorly-addressed/represented via the application of median values, or averages for key parameters, thereby increasing uncertainty and risk. In order to address the fact that slope instability is one of the major sources of risk in open pit mining, largely due to data uncertainties, the discrete fracture network (DFN) and synthetic rock mass (SRM) approaches are applied herein to elucidate the effect that anisotropy has on slope design. The thesis presents the analysis of the interaction of complex geological features with slope geometry, utilising the input from synthetic rock mass modelling and thereby exploiting the integration of structural geology with slope design. The interaction of a geological plane (e.g. foliation, bedding or contact) with a vertical surface within a pit design, results in a calculable apparent dip (AD). Continuous AD maps augment the use of 2D design sections, by providing a rational and dynamic methodology to improve the definition of design sectors, as inputs to the mine planning process, from a very early project stage. The technique shows the spatial variability in the apparent dip of anisotropic units or geological features, using fully-constrained, implicit 3D geological models. Apparent dip is plotted spatially and binned into categories that show its inward or outward dipping attitude with respect to the pit. This binning may be arbitrary or guided by input from SRM analysis or may rely on critical friction angles. This is particularly important where the main plane of anisotropy shows significant changes in dip direction, which may warrant further domain subdivision. This technique is also particularly useful where the pit surface shows a dramatic change in orientation with respect to strike or dip direction in a given part of a domain. Improved design sector classification highlights possible areas of favourable or unfavourable interactions of anisotropic lithologies, with slope geometries in current mining faces, future pushbacks or on final/design faces. This allows for focused data acquisition and practical sector- (or sub-sector)-specific design parameters as input for mine design, in turn leading to optimised designs and early risk mitigation. Utilising the input from the SRM analysis, the new 3D geometrical analysis methodology was successfully applied at the Heuningkranz project in the Northern Cape of South Africa, resulting in the definition of early stage design sectors. A geotechnical block model was developed, constrained with design sectors and populated with geotechnical design parameters as input in the mine planning process. Early stage, focussed geotechnical data acquisition was also planned and implemented based on results obtained from the 3D geometrical analysis. New data were included in the geological and geotechnical databases to inform model updates. The application of the new methodology at the Heuningkranz exploration project showed how early front-end loading of a project, utilising the suggested iterative process, greatly improves the design process to achieve the objectives of the project stage (concept), i.e. to look for fatal flaws and to refine the investigation with new data. From a geotechnical engineering point of view, this was achieved, as focussed data acquisition targets potential risk areas early on, to allow for optimised ramp and waste dump placement, leading to improved slope designs. It therefore increases the confidence in one of the major drivers of this project, viz. the strip ratio, that is vital in the economic viability of the project. A case study at Sishen Mine, in the Northern Cape of South Africa, demonstrates that an ongoing, risk-based approach, described in this thesis, leads to a higher-confidence, dynamically-updatable geotechnical model, thereby allowing for an integrated mine design process that achieves early risk mitigation and at the same time, delivers optimised designs that unlock significant value without compromising the geotechnical risk profile of the mine. Another case study at Kolomela Mine, also in the Northern Cape of South Africa, further validates the approach described in this thesis, that leads to early risk identification and successful mitigation of potential slope instability. Increased confidence in slope design by addressing geological uncertainty led to pit layout optimisation opportunities, unlocking significant value. This was achieved without any fall of ground (FoG) related accidents, due to the early identification and effective management of geotechnical risk through effective collaboration by the Geotechnical Engineering, Mine Planning and Mining Operations Departments. At Jwaneng Mine in Botswana, the application of this new methodology has greatly assisted in the early identification of potential risk areas and allows for forward prediction of stability implications based on still-to-be-mined benches with respect to the mine design. This information leads to a better definition of the design, and early risk mitigation through execution of the solution design (in this case, dip slope mining). Business risks associated with personnel and equipment safety, as well as economic risks, could have had devastating impacts on the mine’s sustainability and, as Jwaneng provides a large part of Botswana’s GDP, a deleterious impact on the country’s economy