Any mining operation is only as good as the information it has and how well it’s organized to extract resources as efficiently as possible. Also important is the presentation of and access to that information. Mining operations that have increased visibility have the ability to make better business decisions around resources, including orebodies. Analysis around consistent raw materials is needed to maintain accurate process control because if the composition of the ore body changes, the extraction process will need to change just as quickly. Traditional mining sample analysis often involves a costly and time-consuming process of sending samples to off-site laboratories and waiting for the results.

Resource Estimation

The Resource Estimation module is used for detailed resource analysis and reporting. It provides a comprehensive range of modelling-related functions. Many of its block modelling tools are also used for mine planning and reserve estimation. Resource estimation is used to determine and define the ore tonnage and grade of a geological deposit, from the developed block model. There are different estimation methods used for different scenarios dependent upon the ore boundaries, geological deposit geometry, grade variability and the amount of time and money available. A typical resource estimation involves the construction of a geological and resource model with data from various sources, depending on the nature of the information and whether the data is hard copy or computerized.

An orebody model serves as the geological basis of all resource estimation, an orebody modelling project starts with a critical review of existing drill hole and surface or underground sample data as well as maps and plans with current geological interpretation. Drill hole and/or sample databases are set up to suit all the quantitative and qualitative information necessary to build a resource model.

Once the geological modelling is completed, the geological envelopes are divided into block models. Subsequently, the estimation of these blocks is done from “composites” that are point measures of the grade of ore in the rock. Several different mathematical methods can be used to do the estimation, depending on the desired degree of precision, quality and quantity of data and of their nature. Models such as resource estimation software can be extremely complex, taking up a large data footprint. Collating and securing the large amounts of data is also as crucial as building the powerful rendering software into the system itself.  This type of technology allows the Geologist to easily and efficiently break the orebody into blocks that can be interacted with and read to give mining geologists and stakeholders the information they require to make smarter business decisions.

Implicit Modelling

Implicit modelling is a game-changing innovation in geological modelling. Traditionally, geological models are produced using a manual drawing process. Sections are defined, and lithologies, faults and veins are drawn on the sections. Lines are then drawn to connect surfaces across multiple sections.

Micromine’s Implicit Modelling module uses radial basis functions (RBFs) to model grade shells, lithology boundaries, faults or surfaces. Using 3D points, polylines, polygons or drill hole intervals as input, it generates wireframe solids representing features like lithology units or zones of a specified grade range, or wireframe surfaces such as fault planes or surface topography. These wireframes are readily displayed in Micromine’s Vizex (Visual Explorer) visualisation environment and are a valuable tool for finalising geological or grade interpretations. Implicit Modelling software is not only gaining mainstream popularity as it is integrated within mining applications, it is believed geologists will use Implicit Modelling to generate grade within the orebodies along with the orebody boundaries.

Frank Bilki, Technical Product Manager at MICROMINE said “Although this is actually possible right now, the results can’t be used for reporting or production planning because the grade values don’t have the necessary supporting data provided by traditional grade estimation methods. Advanced rendering methods like stereoscopic 3D and immersive reality will increasingly be used for all forms of geological modelling.”

Data for Uncertain Ore Bodies

Underground ore bodies are very valuable, yet their shape and size are unknown. In the previous decade, several tactics have been employed for solving geological modelling challenges. This is because the data available is limited and the value of the reserve is very high. With the advancement of technology, new computational methods are being developed which aim at producing an accurate representation of the orebody by optimally using the data. But the basis of modeling approach remains the same with topographical information and drill hole data play a crucial role in helping modelers paint a realistic picture of the orebody.

Their reliable estimation is critical to both the confidence in a feasibility study, and to the day-to-day operation of a mine. Together with sampling, assaying, geological and other errors introduced during interpretation and estimation, additional errors are likely to be introduced during the application of technical and economic parameters used for conversion of resources to reserves. High-quality interpretation and estimation must be supported by high-quality data. Any company expecting to make sound investment or operational decisions must base this on both relevant and reliable information.

An Ore Reserve statement generally contains a single set of grade and tonnage figures without a discussion of the potential inherent errors in these estimates. Some sensitivity studies may be run, but confidence limits are rarely quoted and, if they are, often do not consider many of the factors that cause uncertainty in the grade and tonnage estimates. “Irregularly distributed sample values (drill holes) are extended/distributed to field by several approaches; like triangle, polygon) b. Inverse distance methods, geostatistical methods, artificial intelligence (neural networks)” (Erarslan, 2012).

In the triangular method, triangles are made by joining drill holes. Average grade of the drill holes is found out by multiplying the grades of the triangular areas with the thickness of drillholes. In the polygonal method, polygons are geometrically defined by the perpendicular bisectors of the lines between all points. In the Inverse Distance method, weights are given to points and the weights are inversely related to the distance of drillhole node to the sample point. This is done by taking the weighted average of parametric distances. Generally, the size of the orebody leads to a large and complex database which becomes difficult to handle by hand efforts alone. Then it steps up the numerical algorithm methods and the mathematical approach. This leads to the development of software that increases accuracy and saves time processing.

Emphasis is placed on data quality and the requirement for strong quality management to be linked to continuous improvement.