The application of probabilistic logic to identify, quantify and mitigate the uncertainty inherent to a large surface mining budget

Show simple item record

dc.contributor.advisor Webber-Youngman, R.C.W.
dc.contributor.coadvisor Yadavalli, Venkata S. Sarma
dc.contributor.postgraduate Hager, Johann
dc.date.accessioned 2021-04-30T12:36:10Z
dc.date.available 2021-04-30T12:36:10Z
dc.date.created 2014
dc.date.issued 2014-01
dc.description Dissertation (MSc)--University of Pretoria, 2014. en_ZA
dc.description.abstract Mining is a hugely expensive process and unlike manufacturing is based on an ever diminishing resource. It requires a continuous infusion of capital to sustain production. A myriad of factors, from the volatility of the markets to the surety that the minerals are really there, “plagues” both management and investors. The budget tries to predict or forecast future profits and acts as a roadmap to all stakeholders. Unfortunately, most of the time the budget of a mine degenerates to the extent of a collapse, sometimes very soon into the new budget period. This problem plagues both small and large mines indiscriminately. The budget is dictated in absolutes, and little or no variability is allowed. This thesis aims at developing a process to predict the probability of failure or success through the application of probabilistic logic to the simulation of the budget. To achieve this, a very detailed modelling tool is required. The model must replicate the actual mining process both in time and actual spatial representation. Enabling technology was developed over a period of five years, primarily based on the Runge Software Suite. The use of activity based costing enabled the budget to be simulated and expressed as a probability distribution. A Pareto analysis was done on the main cost drivers to extract the most important elements – or key drivers - that need to be manipulated. These distributions were mapped against real data and approximated with the use of the three parameter Weibull distribution. Simulation using Xeras® (Runge) proved to be impossible. This is due to the time needed for setup and processing. The budget was described as an empirical function of the production tonnages split according to the Pareto analyses. These functions were then utilised in Arena® to build a stochastic simulation model. The individual distributions are being modelled to supply the stochastic drivers for the budget distribution. Income, based on the sales, was added to the model in order for the Nett profit to be reflected as a distribution. This is analysed to determine the probability of meeting the budget. The underlying analysis of an open pit mining process clearly reflects that there are primary variables that may be controlled to trigger major changes in the production process. The most important parameter is the hauling cycle, because the haul trucks are the nexus of the production operation. It is further shown that the budget is primarily influenced by either FTE’s (full time employees, i.e. bodies) or funds (Capex or Opex) or a combination of both. The model uses probabilistic logic and ultimately culminates in the decision of how much money is needed and where it should be applied. This ensures that the probability of achieving the budget is increased in a rational and demonstrable way. The logical question that arises is: “Can something be done to utilise this knowledge and change behaviour of the operators?” This led to (IOPA – Intelligent Operator Performance Analyses) – where the performance or lack thereof is measured on a shift by shift basis. This is evaluated and communicated through automated feedback to the supervisors and operators and is being implemented. Early results and feedback are hugely positive. The last step is prove where capital (or any additional money spend) that is applied to the budget will give the most benefit or have the biggest positive influence on the achievement thereof. The strength of the model application lies therein that it combines stochastic simulation, probability theory, financial budgeting and practical mine schedule to predict (or describe) the event of budget achievement as a probability distribution. The main contribution is a new level of understanding financial risk and or constraints in the budget of a large (open pit) mine. en_ZA
dc.description.availability Unrestricted en_ZA
dc.description.degree MEng en_ZA
dc.description.department Mining Engineering en_ZA
dc.identifier.citation Hager, J 2014, The application of probabilistic logic to identify, quantify and mitigate the uncertainty inherent to a large surface mining budget, MEng Dissertation, University of Pretoria, Pretoria, viewed yymmdd <http://hdl.handle.net/2263/79708> en_ZA
dc.identifier.other M14/9/429 en_ZA
dc.identifier.uri http://hdl.handle.net/2263/79708
dc.language.iso en en_ZA
dc.publisher University of Pretoria
dc.rights © 2019 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.
dc.subject UCTD en_ZA
dc.title The application of probabilistic logic to identify, quantify and mitigate the uncertainty inherent to a large surface mining budget en_ZA
dc.type Dissertation en_ZA


Files in this item

This item appears in the following Collection(s)

Show simple item record