Quantification of Geological Uncertainty and Risk Using Stochastic Simulation and Applications in the Coal Mining Industry S Li 1 , R Dimitrakopoulos 2 , J Scott 3 and D Dunn 4 ABSTRACT Stochastic simulation is a recognised tool for quantifying the spatial distribution of geological uncertainty and risk in earth science and engineering. Metals mining is an area where simulation technologies are extensively used; however, applications in the coal mining industry have been limited. This is particularly due to the lack of a systematic demonstration illustrating the capabilities these techniques have in problem solving in coal mining. This paper presents two broad and technically distinct areas of applications in coal mining. The first deals with the use of simulation in the quantification of uncertainty in coal seam attributes and risk assessment to assist coal resource classification, and drill hole spacing optimisation to meet pre-specified risk levels at a required confidence. The second application presents the use of stochastic simulation in the quantification of fault risk, an area of particular interest to underground coal mining, and documents the performance of the approach. The examples presented demonstrate the advantages and positive contribution stochastic simulation approaches bring to the coal mining industry. INTRODUCTION Coal exploration, mine planning, economic valuation of coal assets, and coal production forecasting depend on the ability to effectively and reliably delineate, understand and assess coal resources and reserves. In turn, this ability supports investment decisions in exploration programs, development and production that are in the order of billions of dollars. Furthermore, Stock Exchange reporting of resources and reserves, aiming to benefit shareholders and attract the investment community, critically depends on the assessment of geological risk. Geological uncertainty is recognised as a critical factor in establishing accurate and reliable estimation, categorisation and economic assessment of coal resources and reserves, in terms of quality and quantity. Incomplete understanding of geological risk, including fault risk, is recognised as a major contributing factor to mining projects not meeting their financial expectations. Stochastic simulation methods offer the technologies used to quantify geological risk. They are increasingly applied for this reason in metal mining and applications are widely reported (Dimitrakopoulos, in press; Dowd, 1997; Ravenscroft, 1992), including several papers in this volume. The practical application of simulation methods has been enhanced with the development of fast and efficient simulation algorithms better enabling the simulation of large, complex orebodies (Benndorf and Dimitrakopoulos, 2007, this volume; Boucher and Dimitrakopoulos, 2007, this volume) and their integration with mine planning, design and production scheduling (Godoy and Dimitrakopoulos, 2004; Ramazan and Dimitrakopoulos, 2007, this volume; Menabde et al, 2007, this volume). When compared to metal mining, there have been limited applications of stochastic simulations in the coal mining industry. Stochastic simulation is now being adopted, recognising the inefficiencies of traditional approaches to: 1. model coal seams based on drill hole information, 2. assign and classify coal resources, 3. establish drill hole spacing requirements for resource classification, and 4. identify the location of faults. Two new developments in modelling geological uncertainty and quantifying the related risk with applications to coal mining are presented herein. The first development, extensively reported in Dimitrakopoulos, Scott and Li (2005), refers to the use of stochastic simulation methods to quantify risk in coal seams estimated with conventional methods, to assist Competent Persons in classifying resources and report the level of error with a given confidence. In addition, the approach developed provides the means to test the performance of drilling patterns and optimise data collection based on the local characteristics of the seam considered and a pre-specified error and confidence level. The second development, detailed in Dimitrakopoulos et al (2001), examines the simulation of fault systems and quantification of fault uncertainty. The performance of the approach in a back analysis study at a mined out part of a longwall coal mine elucidates the method and documents the performance of stochastic modelling, its advantages and characteristics. The methods and work presented in this paper were funded by the Australian Coal Association Research Program (ACARP Projects C7025 and C11042) as well as Anglo Coal Australia, BHP Billiton Mitsubishi Alliance, Coal and Allied (Rio Tinto Coal) and Xstrata (previously MIM). QUANTIFICATION OF GEOLOGICAL UNCERTAINTY AND RISK IN COAL RESOURCE ESTIMATION AND CLASSIFICATION The new JORC Code (2004) requires that resource reporting be related to the level of geological confidence, that is, quantified geological uncertainty, for mining companies listed on the ASX. These companies and their Competent Persons are required to ensure that the resource computations and classifications comply with the basic JORC requirements of transparency, materiality and competency. Traditional approaches to the classification of resource have tended to use subjective criteria to define the limits of measured, indicated and inferred resource polygons. Existing guidelines encourage resource classification based on the maximum distances between drill holes and the number of holes drilled, without sound, scientific justification. The stochastic simulation approach to quantifying errors at a specified confidence interval in coal resource estimation to assist Competent Persons is presented next. Orebody Modelling and Strategic Mine Planning Spectrum Series Volume 14 253 1. CRCMining, The University of Queensland, 2436 Moggill Road, Pinjarra Hills Qld 4069, Australia. Email: s.li@crcmining.com.au 2. MAusIMM, COSMO Laboratory, Department of Mining, Metals and Materials Engineering, McGill University, Frank Dawson Adams Building, Room 107, 3450 University Street, Montreal QC H3A 2A7, Canada. Email: roussos.dimitrakopoulos@mcgill.ca 3. Roche Mining, PO Box 2569, Nerang MDC Qld 4211, Australia. 4. MAusIMM(CP), Manager Geological Services, BHP Billiton Mitsubishi Alliance, GPO Box 1389, Brisbane Qld 4001, Australia. Email: doug.l.dunn@bhpbilliton.com