4
was completed, a comparison was made between the esti-
mated tonnage by model and the tonnage reported at the
crusher by the dispatch system. The mining cuts mined by
bench were employed for this purpose.
• The reserves were used to calculate the tonnage of the
mining cuts by bench.
• The results obtained from each ore control model are
shown in Table 2, including the tonnage obtained
from these mining cuts through the dispatch system.
The last row displays the variation between them.
Based on an analysis of the results, it can be concluded
that the 7.5 × 7.5 × 14 m model exhibits the least variation.
The block size selected was 7.5 × 7.5 × 14, as it resulted in
the least variation in tonnages.
After selecting the appropriate size for the block model,
the ore control model workflow is divided into the follow-
ing steps.
1. Initialize bench.
Define one or multiple benches to be used in the block
model for the update, as well as the bench for filtering the
drillholes for interpolation.
2. Compositing.
The RC were composited to a 14-meter support (bench
length).
3. Reset model items.
Perform a reset on the items involved in the interpolation
process (FeM, FeT, Al2O3, CaO, MgO, P, S, and SiO2), as
well as reset the items that will calculate recoveries.
4. Interpolation of grades.
Interpolation has been carried out using the following steps:
• All items were interpolated from the RC database
(FeM, FeT, Al2O3, CaO, MgO, P, S, and SiO2).
Each item was interpolated in a separate run.
• A generic isotropic variogram (same range in all
directions) with nugget =0, sill =1, and range =
30 m was used.
• A 30 m search range was used to comply with the 10
× 10 grid spacing (3 times the spacing). Interpolation
was performed using a spherical search of 30 m.
• A maximum of 12 composites per block were used,
approximately equivalent to two data circles for each
interpolation point.
• The kriging variance (VAR) was saved only for the
FeM run. This would determine which blocks would
be included in the model.
The kriging variance was calculated based on a var-
iogram and the location of the RC. Since a sill of 1 was
used, the variance would have values between 0 and 1. If a
3D model view is displayed, areas with a variation greater
than 0.6 will clearly show blocks without nearby drill holes.
Therefore, after the interpolation, a calculation is per-
formed to reset all blocks where the variance is greater than
0.6. Figure 7 illustrates Ore control model.
After this, the following process was carried out:
• A calculation adjusted the values of the 8 grade
elements when the variance (VAR) exceeded 0.6.
Additionally, the SG value was set back to a default
value of 2.51, considering FeM=0 using the previ-
ously mentioned formula
• To validate the model, a tonnage-grade curve was
generated for FeM and FeT between assays and the
model, where it can be observed that both follow the
same trend.
Figure 7. Ore control model
Table 2. Tonnage comparison obtained for each model,
compared to the tonnage of the dispatch system
5m x 5m x 14m 2.5m x 2.5m x 14m 7.5m x 7.5m x 14m
Bench Tonnage Tonnage Tonnage
660 63,099 60,747 55,586
842 935,298 935,048 928,185
856 968,491 978,031 945,676
870 1,766,393 1,764,081 1,757,424
884 1,152,273 1,144,678 1,138,934
898 480,185 478,862 484,440
912 1,823,744 1,825,033 1,814,331
926 2,701,473 2,703,083 2,694,727
940 1,316,318 1,316,154 1,305,881
Total 11,207,274 11,205,718 11,125,185
Total (Dispatch
System)
10,781,433 10,781,433 10,781,433
Variation -3.97% -3.96% -3.21%
Block Size
was completed, a comparison was made between the esti-
mated tonnage by model and the tonnage reported at the
crusher by the dispatch system. The mining cuts mined by
bench were employed for this purpose.
• The reserves were used to calculate the tonnage of the
mining cuts by bench.
• The results obtained from each ore control model are
shown in Table 2, including the tonnage obtained
from these mining cuts through the dispatch system.
The last row displays the variation between them.
Based on an analysis of the results, it can be concluded
that the 7.5 × 7.5 × 14 m model exhibits the least variation.
The block size selected was 7.5 × 7.5 × 14, as it resulted in
the least variation in tonnages.
After selecting the appropriate size for the block model,
the ore control model workflow is divided into the follow-
ing steps.
1. Initialize bench.
Define one or multiple benches to be used in the block
model for the update, as well as the bench for filtering the
drillholes for interpolation.
2. Compositing.
The RC were composited to a 14-meter support (bench
length).
3. Reset model items.
Perform a reset on the items involved in the interpolation
process (FeM, FeT, Al2O3, CaO, MgO, P, S, and SiO2), as
well as reset the items that will calculate recoveries.
4. Interpolation of grades.
Interpolation has been carried out using the following steps:
• All items were interpolated from the RC database
(FeM, FeT, Al2O3, CaO, MgO, P, S, and SiO2).
Each item was interpolated in a separate run.
• A generic isotropic variogram (same range in all
directions) with nugget =0, sill =1, and range =
30 m was used.
• A 30 m search range was used to comply with the 10
× 10 grid spacing (3 times the spacing). Interpolation
was performed using a spherical search of 30 m.
• A maximum of 12 composites per block were used,
approximately equivalent to two data circles for each
interpolation point.
• The kriging variance (VAR) was saved only for the
FeM run. This would determine which blocks would
be included in the model.
The kriging variance was calculated based on a var-
iogram and the location of the RC. Since a sill of 1 was
used, the variance would have values between 0 and 1. If a
3D model view is displayed, areas with a variation greater
than 0.6 will clearly show blocks without nearby drill holes.
Therefore, after the interpolation, a calculation is per-
formed to reset all blocks where the variance is greater than
0.6. Figure 7 illustrates Ore control model.
After this, the following process was carried out:
• A calculation adjusted the values of the 8 grade
elements when the variance (VAR) exceeded 0.6.
Additionally, the SG value was set back to a default
value of 2.51, considering FeM=0 using the previ-
ously mentioned formula
• To validate the model, a tonnage-grade curve was
generated for FeM and FeT between assays and the
model, where it can be observed that both follow the
same trend.
Figure 7. Ore control model
Table 2. Tonnage comparison obtained for each model,
compared to the tonnage of the dispatch system
5m x 5m x 14m 2.5m x 2.5m x 14m 7.5m x 7.5m x 14m
Bench Tonnage Tonnage Tonnage
660 63,099 60,747 55,586
842 935,298 935,048 928,185
856 968,491 978,031 945,676
870 1,766,393 1,764,081 1,757,424
884 1,152,273 1,144,678 1,138,934
898 480,185 478,862 484,440
912 1,823,744 1,825,033 1,814,331
926 2,701,473 2,703,083 2,694,727
940 1,316,318 1,316,154 1,305,881
Total 11,207,274 11,205,718 11,125,185
Total (Dispatch
System)
10,781,433 10,781,433 10,781,433
Variation -3.97% -3.96% -3.21%
Block Size