Name three (3) shortcomings of the Conventional models (Estimation)
- DO NOT account for uncertainty. Estimation methods try to approximate some average grade value… not the actual one
- DO. NOT account for local grade variability
- Smoothing/misrepresentation from estimatino on G-T curves has adverse effects on mine plan,design, production forecasts, financial evals, prod. expectations, investments…
Explain this photo
:)
Explain this photo
:)
Discuss this graph
Estimated models (Red) Overestimates low-grades AND Underestimated high-grades, comparatively to simulated models.
Why is Uncertainty ‘not a bad thing’ ?
You can take advantage of uncertainty. Your strategic investments will be sheltered from its adverse effects while remaining exposed to its upside potential. Uncertainty will create opportunities and value.
Develop on this graph
Name four (4) results of Stochastic mine planning
- Lower risk in meeting forecasts
- Higher value for less risk
- Larger ultimate pit limits
- More metal
Case Study: Fimiston Open Pit Mine, Western Australia Orre Boundaries
What are the conclusions?
- Stochastic has higher tonnages for 90% and 10% compared to deterministic
- and
Discuss the general non-linearity of mining ‘transfer functions’
An average type of block model may not provide an average map of the space of response uncertainty; and Nature Does not Know about Distributions Deposit Attributes DO NOT Follow any Given Distribution Model
Discuss the depicted graph
:)
Name two (2) characteristics of Estimation Method (Chap 10)
- An unbiased estimate of a point 𝑍 (𝐱) or volume 𝑍𝑉( 𝐱 ), centred at location 𝐱 ∈ 𝒟, where 𝒟 is the geological domain.
- Estimates smooth-out the low- and high- grades; as a result, they are unable to reproduce the grade distribution and variogram of the input data.
Name two (2) characteristics of Simulation Method (Chap 10)
- Generate a set of possibilities, conditioned to existing data (drillholes, blastholes, etc.) that, as a group, quantify the uncertainty in the deposit.
- More importantly, we wish to reproduce the statistics of the input data (mean, variance, etc.), along with all spatial relationships between points.
Remember this shtuff
Remember this shtuff
