For complex, interconnected plant circuits, engineers utilize the Generalized Least Squares method. This approach minimizes a weighted objective function:
Track the average and range of sub-grouped process variables, such as hourly cyclone overflow densities.
Next came variography: semivariograms, nugget effects, and range. These tools measured how similarity decayed with distance. Lin calculated experimental variograms in multiple directions. The anisotropy was clear: correlation extended farther along strike than down-dip. That mattered for kriging—an interpolator that weights nearby samples according to spatial correlation.
The journey begins at the mine face. Resource estimation, the process of determining if an ore body is economic, relies heavily on geostatistics. Traditional statistical methods assume independence between samples, but ore grades are famously spatially correlated—a high-grade sample is likely surrounded by other high-grade samples. To address this, mineral engineers use . The variogram quantifies how grade variability changes with distance, allowing the engineer to model spatial continuity. This model is then used in kriging , an advanced interpolation technique that provides not only the best linear unbiased estimate of grade in an unsampled block but also a measure of the estimation variance. Without geostatistics, engineers would be guessing at the grade between drill holes, risking either over-capitalization on barren rock or leaving valuable ore in the ground.
The probability of obtaining the observed results if the null hypothesis is true. A p-value below a threshold (typically 0.05) justifies rejecting H0cap H sub 0 Type I Error ( Statistical Methods For Mineral Engineers
If you are working on a specific optimization project, please share you are evaluating (e.g., grinding circuit, flotation cells, leaching pads) and the type of data you are collecting. I can provide a tailored statistical workflow or design an experimental matrix for your exact parameters. Share public link
Statistical methods provide the mathematical framework required to transform noisy plant data into actionable operational insights. By applying these techniques, engineers can quantify uncertainty, optimize recovery rates, and minimize resource waste. 1. Sampling Theory and Variance Control
Y=β0+β1X1+β2X2+…+βnXn+ϵcap Y equals beta sub 0 plus beta sub 1 cap X sub 1 plus beta sub 2 cap X sub 2 plus … plus beta sub n cap X sub n plus epsilon Engineers use the coefficient of determination ( R2cap R squared
Includes over 100 worked examples and downloadable spreadsheets that allow engineers to "flip to the right page" and apply a method to their current plant trial. These tools measured how similarity decayed with distance
: Understanding how measurement errors from assays and sampling impact your conclusions.
When the mining company announced the new high-grade deposit at Cerro Viento, the regional team called her in. The deposit’s assay data were messy: clusters of high values, long tails of low-grade samples, and pockets where grade rose and fell with little warning. Investors wanted a single confident estimate of recoverable metal. The foreman wanted a drill plan. Politicians wanted reassurance that the mine wouldn’t poison the groundwater. And Amaya wanted to teach her students one more lesson — that sound decisions begin where curiosity collides with uncertainty.
For the modern mineral engineer, statistics is more than just math—it is a risk-management tool. By moving from "gut feeling" to data-driven decision-making, engineers can reduce waste, improve environmental outcomes, and ensure the economic viability of mining projects.
Used primarily in reliability engineering to model the breakdown rates of liners, lifters, and crusher components. 2. Sampling Theory and Error Minimization With the models fitted
): Screen a large number of factors (e.g., impeller speed, solids concentration, gas holdup) to identify which ones significantly impact recovery.
Modern practice uses weighted least squares, where each measurement is assigned a variance (from sampling and analytical error). Measurements with low variance receive small adjustments; bad actors receive large adjustments—flagging them for review.
They built nested variogram models: a small nugget to capture sampling and microscale variability, a short-range spherical structure for pocket-scale continuity, and a longer-range exponential structure for broad-grade trends. With the models fitted, ordinary kriging produced smoothed grade estimates across the block model, but Amaya knew kriging’s smoothing bias could underestimate high-grade variability — dangerous for resource classification and project economics.