5
through engineering design practices, which often rely on
empirical factors and allowances to compensate for lim-
ited data and uncertainty. Additionally, comminution or
throughput risk can be managed using resource assay data
to predict changes in lithology and alteration that may
impact on plant throughput.
Comminution risk management involves using estab-
lished design methods and applying allowances and contin-
gencies that result in additional capital expenditure (Bailey
and Lane et al., 2009). For instance, projects may “over-
size” mills or incorporate design redundancies to account
for variability in ore hardness, as with the Constancia proj-
ect (Lane et al., 2015). These strategies ensure operational
stability, even when limited comminution test data are
available, by prioritizing robust circuit designs over precise
throughput forecasting during the early project stages. The
advantage of this approach is that, provided the mine can
cater for increased plant throughput, the project can use
the additional plant capacity to increase NPV above study
projections.
An important late phase of feasibility study assess-
ment should involve evaluating the design against the cur-
rent geometallurgical model status. Often, process design
criteria and subsequent equipment selection are based on
prefeasibility study-level data. Further sampling, testing,
and geometallurgical modelling are conducted to refine
performance forecasting during the feasibility study period.
At the completion of this phase of work, the value of a proj-
ect can be significantly improved based on effective sample
selection and comminution test work conducted during the
feasibility study.
The focusing question is, “When do you know you
have tested enough samples to manage risk?” This should
be a cost-benefit equation, with the inputs being the cost of
test work and time required vs. the improved NPV based
on reduced margins and improved confidence in ore vari-
ability prediction. As testing samples becomes cheaper
(without losing confidence in the quality of the data), more
samples should be tested.
Insufficient comminution testing compromises the
accuracy of mill throughput estimates (Bueno 2015 and
Lara 2024), which are critical for Net Present Value (NPV)
calculations and can affect project valuation. This under-
scores the essential role of comprehensive comminution
testing in order to create robust financial forecasts.
Figure 7 illustrates how changes in throughput impact
the sensitivity of Net Present Value (NPV) across projects.
In this analysis, which focuses on 12 projects out of 154
reviewed feasibility studies, project J emerged as the most
negatively affected: a 25% reduction in throughput caused
its NPV to decrease by over 90%. On the other hand, the
project G exhibited the highest positive sensitivity, with its
NPV doubling in response to a 10% increase in revenue,
influenced by both throughput and recovery rates.
The data from the graph also shows that, 1% change
in throughput corresponds to a 3.36% change in median
NPV. However, when the two most sensitive projects—
Projects G and K—are excluded, the median NPV change
per 1% throughput variation reduces to 2.87% median
NPV.
Table 2 lists the projects plotted in Figure 7, detailing
the country of each project in the feasibility study and the
primary commodity targeted for exploration. Consistent
with the trend shown in Figure 1, most of these projects
focus on gold, with Canada having the highest number of
projects represented in Table 2.
Even a 1% variation in throughput can drive a more
than 4% change in NPV, as seen in Figure 7. This empha-
sizes the importance of thorough orebody geometallur-
gical characterisation to improve throughput forecasts,
enabling projects to achieve more predictable financial
returns. Conversely, inadequate throughput forecasting can
introduce unforeseen variability, potentially undermining
a project’s economic viability as most mining operations’
performance is tied to their grinding circuit performance.
(Lara et al., 2024).
We propose a progressive and adaptive testing approach
to address the critical question of how many samples are
sufficient for reliable throughput forecasts and risk manage-
ment. This method begins with an initial phase of broad,
lower-cost tests -e.g., geochemical assays or proxy tests like
point load, HIT (Kojovic, 2016) or Geopyörä breakage
Table 2. Used projects from Figure 7
Project Country Commodity
Capex (M
US$)
A Argentina Lithium 268.9
B DRC Tin 119.2
C USA Aluminium 108.2
D Mali Gold 196.3
E Australia Scandium 87.1
F Canada Zinc 252.9
G Sudan Gold 328.8
H Brazil Gold 145.2
I Canada Gold 109.8
J Canada Copper 352.0
K Côte d’Ivoire Gold 493.3
through engineering design practices, which often rely on
empirical factors and allowances to compensate for lim-
ited data and uncertainty. Additionally, comminution or
throughput risk can be managed using resource assay data
to predict changes in lithology and alteration that may
impact on plant throughput.
Comminution risk management involves using estab-
lished design methods and applying allowances and contin-
gencies that result in additional capital expenditure (Bailey
and Lane et al., 2009). For instance, projects may “over-
size” mills or incorporate design redundancies to account
for variability in ore hardness, as with the Constancia proj-
ect (Lane et al., 2015). These strategies ensure operational
stability, even when limited comminution test data are
available, by prioritizing robust circuit designs over precise
throughput forecasting during the early project stages. The
advantage of this approach is that, provided the mine can
cater for increased plant throughput, the project can use
the additional plant capacity to increase NPV above study
projections.
An important late phase of feasibility study assess-
ment should involve evaluating the design against the cur-
rent geometallurgical model status. Often, process design
criteria and subsequent equipment selection are based on
prefeasibility study-level data. Further sampling, testing,
and geometallurgical modelling are conducted to refine
performance forecasting during the feasibility study period.
At the completion of this phase of work, the value of a proj-
ect can be significantly improved based on effective sample
selection and comminution test work conducted during the
feasibility study.
The focusing question is, “When do you know you
have tested enough samples to manage risk?” This should
be a cost-benefit equation, with the inputs being the cost of
test work and time required vs. the improved NPV based
on reduced margins and improved confidence in ore vari-
ability prediction. As testing samples becomes cheaper
(without losing confidence in the quality of the data), more
samples should be tested.
Insufficient comminution testing compromises the
accuracy of mill throughput estimates (Bueno 2015 and
Lara 2024), which are critical for Net Present Value (NPV)
calculations and can affect project valuation. This under-
scores the essential role of comprehensive comminution
testing in order to create robust financial forecasts.
Figure 7 illustrates how changes in throughput impact
the sensitivity of Net Present Value (NPV) across projects.
In this analysis, which focuses on 12 projects out of 154
reviewed feasibility studies, project J emerged as the most
negatively affected: a 25% reduction in throughput caused
its NPV to decrease by over 90%. On the other hand, the
project G exhibited the highest positive sensitivity, with its
NPV doubling in response to a 10% increase in revenue,
influenced by both throughput and recovery rates.
The data from the graph also shows that, 1% change
in throughput corresponds to a 3.36% change in median
NPV. However, when the two most sensitive projects—
Projects G and K—are excluded, the median NPV change
per 1% throughput variation reduces to 2.87% median
NPV.
Table 2 lists the projects plotted in Figure 7, detailing
the country of each project in the feasibility study and the
primary commodity targeted for exploration. Consistent
with the trend shown in Figure 1, most of these projects
focus on gold, with Canada having the highest number of
projects represented in Table 2.
Even a 1% variation in throughput can drive a more
than 4% change in NPV, as seen in Figure 7. This empha-
sizes the importance of thorough orebody geometallur-
gical characterisation to improve throughput forecasts,
enabling projects to achieve more predictable financial
returns. Conversely, inadequate throughput forecasting can
introduce unforeseen variability, potentially undermining
a project’s economic viability as most mining operations’
performance is tied to their grinding circuit performance.
(Lara et al., 2024).
We propose a progressive and adaptive testing approach
to address the critical question of how many samples are
sufficient for reliable throughput forecasts and risk manage-
ment. This method begins with an initial phase of broad,
lower-cost tests -e.g., geochemical assays or proxy tests like
point load, HIT (Kojovic, 2016) or Geopyörä breakage
Table 2. Used projects from Figure 7
Project Country Commodity
Capex (M
US$)
A Argentina Lithium 268.9
B DRC Tin 119.2
C USA Aluminium 108.2
D Mali Gold 196.3
E Australia Scandium 87.1
F Canada Zinc 252.9
G Sudan Gold 328.8
H Brazil Gold 145.2
I Canada Gold 109.8
J Canada Copper 352.0
K Côte d’Ivoire Gold 493.3