Statistical Methods For Mineral Engineers ~repack~ -

F⋅f=C⋅c+T⋅tcap F center dot f equals cap C center dot c plus cap T center dot t

It is considered a standard reference text for plant metallurgists and assay chemists to translate vague observations into demonstrable facts. like regression modeling or experimental design in more detail?

Total Sampling Error (TSE) consists of several distinct mathematical components:

p-value = 0.003 (<0.05). Reject H₀.

Once DoE has identified the critical factors, RSM is a collection of mathematical and statistical techniques used to model and optimize the response. In the context of flotation, RSM would create a regression model relating the input factors (e.g., frother dosage, air flow rate) to the output responses (e.g., copper recovery, concentrate grade). The goal is to find the combination of factors that maximizes a desired response, such as economic recovery.

: Moving beyond "gut feeling" to using statistical tools (many of which are built directly into Excel ) to prove whether a process change truly improves recovery or throughput. Key Topics Covered

Once the variogram has been modeled, the next step is to use it to perform spatial interpolation through a process called . Named after the South African mining engineer Danie Krige, Kriging is a generalized linear regression method that provides Best Linear Unbiased Estimates (BLUE) . This means it minimizes the variance of the estimation error (the "kriging variance"). Statistical Methods For Mineral Engineers

Factorial Design Matrix (2^3 Example) --------------------------------------------- Run | Factor A (pH) | Factor B (Collector) | Factor C (Air) --------------------------------------------- 1 | - | - | - 2 | + | - | - 3 | - | + | - 4 | + | + | - 5 | - | - | + 6 | + | - | + 7 | - | + | + 8 | + | + | + --------------------------------------------- Factorial Designs 2k2 to the k-th power factorial design evaluates factors at two levels: high (+) and low (-). For example, a

“People will want averages,” Lin said. “But the mean will be dragged by those outliers. If we present that, we’re lying by decimal point.”

Gy's theory breaks down total sampling error into components. The most significant is the Fundamental Sampling Error (FSE) , which is inversely proportional to sample mass and directly proportional to the size of the largest particles in the ore (the "liberation size"). By identifying and minimizing these components, engineers can design protocols that define the minimum sample mass required for a given particle size distribution to ensure a statistically representative sample. This is especially critical for the "nuggety" gold deposits where a single grain can dominate the assay value. F⋅f=C⋅c+T⋅tcap F center dot f equals cap C

This is the specific branch of statistics developed for the mining industry to estimate reserves based on sparse drill hole data.

: Using the seven-step process to draw conclusions about process changes.

They tested for normality and quickly rejected it. The grade distribution was log-normal with heavy tails. Amaya suggested a log-transform for many analyses but warned against blind application. “Transformations help with modeling, not with telling the whole story,” she said. “We have to interpret back in original units for engineering decisions.” Reject H₀

The minimum unavoidable error resulting from the constitutional heterogeneity of the material (e.g., the fact that valuable minerals are discrete grains locked inside waste rock). FSE can only be reduced by crushing the sample to a smaller particle size before splitting.

For too long, mineral engineers relied on rules of thumb: “Take a cut every hour,” “Double the sample if in doubt,” “The lab must be wrong.”