Mine-to-Mill 2.0: Turning Grade Volatility into Predictable Cash Flow in Gold, Silver, Copper and Iron Operations

For decades, the promise of "Mine-to-Mill" optimization has been clear: integrate the entire value chain from the geology of the pit to the final product to maximize throughput and recovery. Yet, for many operations, this promise remains unfulfilled. Traditional Mine-to-Mill 1.0 initiatives often stall at blast-to-grind optimization, focusing heavily on fragmentation and energy efficiency while leaving critical gaps in stockpile management and plant feed predictability.

The result is a disconnected value chain where geology, mine planning, and metallurgy operate in silos. Information is lost at every handover—from the pit to the truck, to the stockpile, and finally to the plant. This loss of fidelity turns ore variability into an unmanaged risk, leading to reactive plant control, lost recovery, and inflated operational costs.

The industry is now evolving toward Mine-to-Mill 2.0. This new paradigm moves beyond simple fragmentation models to a unified, data-driven framework that connects geology, planning, and operations. By leveraging advanced AI and digital twin technology, operations can now predict, manage, and turn ore variability into a controllable advantage.

The Core Problem: Hidden Variability and Information Loss

The fundamental challenge in modern mining isn't just declining grades; it is the inability to see and manage variability as material moves downstream.

In a typical operation, a high-resolution block model exists in the mine planning software. However, as soon as that ore is blasted and loaded into a truck, the rich data associated with it is often simplified into a polygon-averaged grade. And once the truck dumps this material onto a stockpile, it is further aggregated into a coarse average.

Reclamation planners operate on these coarse averages for grade and other ore properties. And by the time the material is reclaimed and fed to the processing plant, the operators are flying blind. They are managing a "black box" feed based on rough averages that mask critical fluctuations in grade, hardness, and mineralogy.

This information loss creates a ripple effect of inefficiency:

• At the Mine: Ore is loaded with limited visibility into its true properties, leading to misclassification and incorrect routing.
• At the Stockpile: Variability is homogenized in data systems but remains physically heterogeneous, leading to poor blending decisions.
• At the Plant: Operators react to changes in feed characteristics only after they occur, often too late to prevent recovery losses or throughput drops.

The Cost of Variability

When variability is unmanaged, the plant pays the price in three distinct ways:

1. Lost Recovery: When feed grade or mineralogy shifts unexpectedly, the plant cannot adjust reagent dosage or retention times fast enough. For example, a sudden spike in high-grade ore without a corresponding increase in cyanide dosage results in gold being lost to tailings.
2. Reduced Throughput: Unexpected changes in hardness or clay content force the mill to slow down to maintain stability, reducing overall production capacity.
3. Inflated Operational Costs: To buffer against uncertainty, operators often run conservatively, overdosing reagents or under-utilizing mill power. This leads to excessive consumption of energy and consumables like cyanide or grinding media.

The financial impact is significant. For a mid-tier gold producer, unmanaged variability can cost millions annually in lost revenue and wasted reagents.

Introducing Mine-to-Mill 2.0: A Unified Framework

Mine-to-Mill 2.0 solves this disconnect by creating a continuous digital thread from the resource model to the plant control system. It is built on three pillars: Track, Plan, and Optimize.

• Track: Preserve granular ore properties (grade, hardness, mineralogy) from the pit through to the stockpile and ROM pad, ensuring no data is lost during material handling.
• Plan: Forecast future feed characteristics and operational outcomes (throughput, recovery) using geometallurgical models, allowing teams to adjust strategies proactively.
• Optimize: Operationalize these insights to make the best decisions regarding blending, routing, and plant control setpoints.

The NTWIST MineMax Suite

NTWIST’s MineMaxsuite embodies this 2.0 framework, offering specific modules that address each stage of the value chain.

OreMax: Stockpile Resource Modeling

OreMax transforms stockpiles from black boxes into high-confidence assets. It tracks material from the pit to long term stockpiles with full auditability, creating detailed block models. By replacing averages with granular distributions, it allows geologists to quantify stockpile resources and provides ammunition to planners to convert stockpile resources into reserves.

DynaMax: Short-Term Stockpile Modeling & ROM Pad Visibility

DynaMax provides real-time visibility of the ROM pad. It replaces finger averages with block-level precision, tracking ore properties as they sit or move on the pad. This allows for accurate short-term feed forecasting, ensuring that planners know exactly what material will be fed to the plant.

PlanMax: Blending & Reclamation Optimization

PlanMax uses the inventory data from OreMax and DynaMax to build optimal blending strategies. It aligns extraction plans with plant constraints, ensuring a consistent feed that meets production targets while minimizing downstream variability.

MillMax: Real-Time Plant Control & Optimization

MillMax acts as the final control loop. It uses predictive geometallurgical models and soft sensors to forecast how the current feed will behave in the plant. It then provides prescriptive recommendations for setpoints—such as reagent dosage, air rates, or pump speeds—to optimize recovery and throughput in real-time.

Industry Applications & Proof in Practice

The shift to Mine-to-Mill 2.0 is already delivering measurable value across various commodities.

Gold: Recovery Improvement and Reagent Savings

At a 100 koz/year gold operation, the integration of DynaMax and MillMax provided real-time visibility into feed variability. The system predicted the impact of feed changes on recovery and recommended optimal SIBX (collector) dosages.

• Result: The operation achieved a 1.6% increase in recovery and a 30% reduction in reagent costs by moving from static dosing to predictive control.

Gold: Preventing Ore Misplacement

At an open-pit gold mine in Brazil, ramp-up pressures led to frequent routing errors where high-grade ore was sent to waste. NTWIST implemented a tracking system that visualized these errors.

• Result: Misrouting decreased by 98%, preventing an estimated $0.5M in annual ore losses. Additionally, OreMax identified 100kt of previously unaccounted medium- and high-grade ore valued at US$11M in a low-grade stockpile.

Gold/Copper: De-risking Stockpile-Only Operations

A gold-copper mine prepared to transition to a stockpile-only operation. Performing accurate reserve estimation was critical. NTWIST deployed a probabilistic model that mapped grade and confidence to each stockpile block.

• Result: The model validated the site estimate with high precision (Copper within -1.7%, Gold within +0.7%) and provided a confidence map that allowed for risk-aware planning and blending.

Transforming Operations in 90 Days

Implementing Mine-to-Mill 2.0 does not require a multi-year digital transformation overhaul. NTWIST utilizes a 90-day blueprint to deliver quick wins:

1. Weeks 1-2: Assess data availability and establish baselines.
2. Weeks 3-6: Deploy digital twins for stockpiles (OreMax/DynaMax) to visualize inventory.
3. Weeks 7-10: Calibrate predictive models for throughput and recovery.
4. Weeks 11+: Activate optimization strategies for blending and plant control.

This rapid deployment allows operations to start realizing value—such as improved feed predictability and reduced misplacement—within a single quarter.

Conclusion

The mining industry can no longer afford to let valuable data evaporate between the mine and the mill. Mine-to-Mill 2.0 represents a necessary evolution in operational strategy, turning the challenge of ore variability into a source of predictable cash flow.

By connecting geology, planning, and processing through a unified, AI-driven framework, NTWIST empowers operations to see what is hidden, predict what is coming, and optimize every ton that enters the plant.

Ready to unlock hidden value?

Discover how much value is hiding in your stockpiles and process data. Book a complimentary assessment with NTWIST today to analyze your current performance and identify immediate opportunities for optimization.

Get Your Free Stockpile Assessment