Which blocks should be mined first? When should they be extracted? And where should the material go?
In this K-MINE webinar clip, discover how genetic algorithms tackle these decisions through a two-stage optimization process—connecting pit optimization and long-term mine scheduling with ore blending, processing plants, and stockpile allocation.
Video transcription
How Genetic Algorithms Work in Mine Planning
What is the main principle of our genetic algorithms?
We use a two-step approach.
The first step is to define the amount of mining material that should be extracted. We're looking for the optimal amount based on our input parameters.
Block Selection and Long-Term Mine Scheduling
In the pit optimization module, we can select a group of blocks and use it as a constraint.
If we want to create a life-of-mine plan, we can use all the blocks within the selected pit. The life-of-mine result from pit optimization becomes one of our constraints.
Genetic algorithms are intended for long-term planning because they work with groups of blocks. Using this approach for a short-term plan would not give the expected result.
Once the required group of blocks is defined, the algorithm determines when each block should be mined.
It creates the mining schedule for those blocks.
Optimizing Ore Blending and Material Destinations
Then the second-stage algorithm determines the optimal blending and the destination for each block.
Should the material go to a processing plant or a stockpile?
If it goes to a stockpile, which one?
That's what the algorithm determines.