Modelling the impact of change in systems and technology in a surface mining environment with system dynamics

Abstract

Methods for estimating the impact of disruptive technologies in a surface mining environment are explored that will enable mining managers to justify the introduced technology. This was dependent on the estimation of time-based productivity and costs. Each sub process in mining operations works toward a long-term goal of minimized costs and maximized production. These sub processes are intricately dependent on each other. Due to high inter-dependency of these processes the impact is not easily captured for the total mine with traditional spread sheet modelling. The dynamic nature of such a mining environment required a dynamic modelling approach. The analysis required all relevant criteria that can be obtained in a mine to be defined and was represented either as raw data inputs or as statistical parameters describing the behaviour of such data. The dynamic modelling heavily depended on these statistical parameters, since using raw data would further complicate any predictive analysis and modelling. In addition, cyclic mining processes and dependencies required a high-level description of the “as is” situations. The research outcomes were dependent on a good understanding of variability of the processes that are normally due to the limitations of the mining machinery, the quality of the work impacted by people interacting with such machinery, mine structure and the sequence of the unit processes. The simulation also depended on many attributes, constants, auxiliaries and tables of data to describe behaviour from typical mining operations in similar geotechnical environments. These are defined adequately to model the mining environment. The steps followed in the thesis are based on a top-to-bottom-to-top dynamic systems modelling method (DSM) as discussed in Chapter 2.The system dynamics simulation (SD) tool used is called Vensim, a widely used SD modelling tool. The justification of the method as well as ways of formulating the model structure required understanding of the system dynamics modelling and quantification techniques and is discussed in detail in chapter three. Most of the inputs used in the creation of the model comes from mining engineering literature of the typical input parameters and user experience based inputs. It required review of a large number of publications including system dynamics modelling based mining literature. This was discussed in detail in chapter four. The technique of testing the validity of the model is explained in chapter 5. The model testing methods were implemented from start to finish as the system dynamics model was built from the simple to the complex. The total mining value chain of a typical surface mine was discussed in detail to determine all quantifiable endogenous or exogenous variables. The causalities and relationship are summarized to form the conceptual model initially, then the model is refined using a specific case study identified. The identified case study was also used to test the model as discussed in chapter six with appropriate testing methods described in chapter five. Chapter 6 also explains steps taken towards creating the larger total mine model by modelling chunks of sub-processes which are drilling, blasting, loading, hauling and crushing. Finally, application of mine specific data and discussion of the results are discussed in chapters seven and eight. The thesis has shown that it is possible to represent the total mine value chain in a simulation modelling environment using system dynamics tools effectively at the abstract level to find answers to specific mining problems. The created model is generic enough so that it may apply to ` size and shape of the surface mining environment. The model created provides answers to the research questions in sufficient detail in terms of financials and efficiencies as the model is able to provide a quantified answer. The approach to quantification of impact of drill automation is based on the elimination of drilling deviation. The main assumption is that automation will lessen the variation due to resource limitations related to low quality drilling and blasting parameters and also drilling deviation related inefficiencies, by achieving the targeted or planned quantities of drilling required to achieve annual tons of ore production. The benefit of reducing the drill deviation by about 10% is calculated as 610 million Rands per year for the simulated case study which has a simulated mining cost of 16.1 billion Rands per year excluding labour and overheads. This amount easily justifies the automation of all the drill rigs in a mine. There could be other hidden benefits due to less variation in the mining environment due to on-time drilling being achieved which is beyond the scope of this thesis. In conclusion, a novel parametric system dynamics model helps mining engineers to get quick answers to systematic changes made to mine excavation related parameters and the overall impact of these changes to the cost of mining as well as any efficiencies. The drilling quality is a significant parameter in any mine. The consequences of not following the designed drilling parameters lead to many complications. The research which resulted in a detailed system dynamics simulation model can give quick answers to mining engineers in terms of expected fragmentation levels, changes to resource requirements in terms of drilling, blasting, loading, hauling and their related costs. The technological changes which force the mine to change the planning of the mine environment on a daily basis also causes disruptions to the other mining processes. Therefore, SD modelling helps the management to have a deep understanding of the interactions of key parameters.