No statistical method for mineral engineers is complete without addressing the fundamental error of sampling.
I can help identify the best statistical approach for your data.
This error is inherent to the material's mineralogical composition and particle size distribution. It is the only error that cannot be eliminated, though it can be reduced by crushing the material to a smaller top size before splitting.
Traditional statistical process control (SPC) charts assume that observations are independent and identically distributed – an assumption that rarely holds in mineral processing. Pierre Gy and his followers developed a more appropriate framework called , which applies variographic analysis to time‑ordered process data. The chronostatistical variogram decomposes process variability into components that can be assigned to different sources (feed heterogeneity, instrument noise, control loop instability), allowing the engineer to pinpoint structural problems that conventional control charts miss. As practitioners have observed, if a control chart shows a process is out of control, it must also suggest logical directions for solutions – a principle that chronostatistics fulfils elegantly. Statistical Methods For Mineral Engineers
To help apply these concepts to your specific operation, could you share a bit more context?
Geostatistics is a specialized branch of statistics used for spatial modeling.
Process engineers model the impact of operational changes on performance. For instance, analyzing how froth height changes affect the kinetic constant ( ) of flotation requires linear regression and plotting diagrams to determine the collection zone efficiency. 2.5. Multivariate Techniques Mining data is rarely univariate. No statistical method for mineral engineers is complete
Minimize risks during scale-up from bench-scale tests to full production. 2. Sampling Theory and Error Mitigation
Using histograms and box plots to identify outliers in sampling data, ensuring data integrity. 2.2. Error Minimization and Mass Balancing
The optimization algorithm minimizes the sum of squared adjustments, delivering a mathematically sound mass balance that reflects the most statistically likely state of the plant. 5. Hypothesis Testing in Plant Trials It is the only error that cannot be
Mineral engineers frequently evaluate whether an operational change—such as a new frother formula, a change in cyclone pressure, or a new grinding media blend—actually improved performance. Hypothesis testing removes subjectivity from these decisions. Common Statistical Tests in the Plant
factors at two levels (high and low). For example, testing 3 factors (pH, collector dosage, and air flow rate) requires distinct runs. This maps all main effects and interactions.
Caused by the spatial distribution of minerals (e.g., heavy minerals settling to the bottom of a conveyor belt). GSE is minimized by taking many small increments over time rather than one large grab sample. Statistical Quality Control (SQC)
—to minimize the Fundamental Sampling Error (FSE). By applying variance analysis, engineers determine the minimum sample mass required to represent the larger lot, ensuring that downstream decisions aren't based on skewed data. 2. Process Optimization and Design of Experiments (DoE)