value below 1.0 indicates the process is generating out-of-specification material. 8. Summary of Statistical Tools for Mineral Engineers Statistical Tool Primary Application Area Primary Benefit Baseline Plant Assessment
Mineral engineering deals with heterogeneous materials. Unlike manufacturing industries where raw inputs are highly standardized, a mineral processing plant treats run-of-mine (ROM) ore that varies continuously in mineralogy, hardness, and feed grade.
Analyzing experimental data to find the optimal reagent blend that maximizes recovery without wasting material. 2.4. Statistical Analysis of Operational Parameters Statistical Methods For Mineral Engineers
While kriging provides a “best” estimate, it smooths local grade variations and underestimates the full range of possible outcomes. Conditional simulation (also called stochastic simulation) overcomes this limitation by generating multiple equally probable realisations of the deposit – each honouring the sample data and the variogram model. The ensemble of realisations directly quantifies spatial uncertainty and can be used for risk‑based mine planning, strategic selection, and grade‑tonnage curve analysis. High‑order simulation techniques, which do not assume multivariate Gaussian distributions, can reproduce complex geological patterns characteristic of many ore deposits.
Evaluates categorical data, such as comparing the frequency of specific equipment failure modes across different shifting crews. Key Concepts in Testing Null Hypothesis ( H0cap H sub 0 value below 1
Geometallurgy is the discipline that links geological characteristics of the orebody to processing performance. Statistical methods lie at its core.
Compares the means of two groups. A paired t-test evaluates the same circuit before and after a specific change (e.g., changing a frother type). An independent t-test compares two parallel flotation banks running different reagents. Unlike manufacturing industries where raw inputs are highly
F⋅f=C⋅c+T⋅tcap F center dot f equals cap C center dot c plus cap T center dot t
charts): Track process averages and variability over time to detect shifts caused by equipment wear, changes in ore hardness, or operator errors.
) to keep the relative variance of the fundamental sampling error ( σFSE2sigma sub cap F cap S cap E end-sub squared
A popular DoE technique used to model the relationship between multiple input variables (e.g., reagent dosage, collector type) and output responses (grade, recovery).