Calculator D3

Grade Reconciliation Workflow: From Blasthole Assay to Mill Feed

Grade reconciliation is how mining and processing teams compare what the rock *should* contain (based on blasthole assays) with what the mill *actually receives*, so they can adjust blasting, blending, or circuit settings in real time.

Typical Scale
Operates at 10–50 blast rounds/week; reconciles 50k–500k tonnes/round
Industry Standards
JORC Code (Section 20), CIM Definition Standards (2022), ISO 17025 (assay labs)
Real-Time Threshold
Grade deviation > ±0.10% Cu or ±0.3 g/t Au triggers immediate blend override protocol

⚠️ Why It Matters

1
Inconsistent blasthole assay sampling
2
Biased grade estimation
3
Incorrect ore/waste delineation
4
Suboptimal blending decisions
5
Mill feed grade deviation > ±0.15% Cu
6
Increased regrind duty, penalty metal recovery loss, and circuit instability

📘 Definition

Grade reconciliation workflow is a closed-loop engineering process that quantitatively links blasthole assay data—collected during drill-and-blast operations—to mill feed grade predictions and actual plant throughput measurements. It integrates geostatistical modeling, sampling theory, mass balancing, and real-time process analytics to manage ore variability, enforce grade control tolerances, and enable mine-to-mill optimization. The workflow relies on rigorous QA/QC protocols, spatially anchored sample traceability, and dynamic feedback to upstream blasting and haulage decisions.

🎨 Concept Diagram

BlastholeDrill & SampleLab AssayReconciliationFeedback LoopCircuit Adjust(SAG Speed, Blend)

AI-generated illustration for visual understanding

💡 Engineering Insight

Grade reconciliation isn’t about ‘matching numbers’—it’s about diagnosing *where uncertainty enters the value chain*. A 0.05% Cu discrepancy rarely stems from assay error alone; it’s usually the visible tip of misaligned sampling protocols, unmodelled geological domains, or uncalibrated mass measurement. The most effective reconciliation programs treat each deviation as a forensic signal—not a statistical outlier—and trace it backward through the workflow until the engineering root cause is isolated and corrected.

📖 Detailed Explanation

At its foundation, grade reconciliation starts with the physical act of sampling: collecting representative fragments from blastholes drilled through heterogeneous rock. Each sample must preserve spatial context (depth, collar location, orientation) and be processed under strict QA/QC to avoid contamination or dilution—because even a 2% moisture error in chip sampling can shift Cu grade by 0.02% in oxide ores.

Beyond sampling, reconciliation requires rigorous mass balancing: every tonne assigned to a blast round must be physically tracked via haul truck GPS, weighbridge logs, and stockpile survey data. This demands integration between mine planning software (e.g., MineSuite, Deswik), laboratory LIMS, and plant DCS systems. Discrepancies here often reveal systemic gaps—such as unrecorded dozer pushbacks or undocumented stockpile rehandling—that no assay can resolve.

At the advanced level, modern reconciliation leverages digital twin frameworks: feeding blasthole assay grids, real-time conveyor belt mass flow, and online elemental analyzers into a dynamic grade prediction engine. These engines apply conditional simulation (e.g., Sequential Gaussian Simulation) to quantify grade uncertainty envelopes—not just point estimates—and trigger adaptive control actions only when deviations exceed statistically justified thresholds (e.g., p < 0.01 for grade shift detection). This transforms reconciliation from retrospective reporting into anticipatory process governance.

🔄 Engineering Workflow

Step 1
Step 1: Blasthole drilling & georeferenced sample collection (core/RC chips with depth tags)
Step 2
Step 2: Laboratory assay (ICP-MS or AA) with certified blanks/duplicates and CRM validation
Step 3
Step 3: Grade compositing & spatial attribution to blast round and bench/block ID
Step 4
Step 4: Mass reconciliation: match assay-weighted grade to measured tonnage hauled and stockpiled
Step 5
Step 5: Mill feed prediction vs. real-time online analyzer (e.g., PGNAA) or composite lab assay
Step 6
Step 6: Deviation analysis (grade, tonnage, metal content) and root-cause classification (sampling, blending, haulage, assay)
Step 7
Step 7: Closed-loop action: update blast design parameters, adjust blend ratios, or modify circuit setpoints (e.g., SAG speed, cyclone pressure)

📋 Decision Guide

Rock/Field Condition Recommended Design Action
Assay bias factor < 0.94 AND compositing interval > 1.5 m Revert to 0.75-m compositing; implement duplicate core-drill QA checks; recalibrate assay database using certified reference materials
Mass pull error > ±5.0% AND assay-to-feed lag > 96 hrs Install real-time belt scale + density meter on primary crusher feed; deploy portable XRF at stockpile reclaim point
Grade reconciliation variance > 0.08% Cu (std dev) across 10 consecutive blasts Trigger geostatistical review: test variogram anisotropy, reassess search ellipsoid parameters, verify domain boundaries against structural mapping

📊 Key Properties & Parameters

Assay Sample Density

1.2–3.5 assays/tonne for copper porphyry; 0.8–2.0 for gold oxide

Number of blasthole assays per unit volume of blasted rock (e.g., per 10,000 t or per blast round)

⚡ Engineering Impact:

Directly governs confidence in block model grade estimates and detection sensitivity for high-grade outliers

Compositing Interval

0.5–2.0 m (commonly 1.0 m for hard-rock porphyry, 0.75 m for soft sedimentary ores)

Vertical length over which blasthole assay segments are averaged to represent a single ‘sample’ for reconciliation

⚡ Engineering Impact:

Too coarse → smears grade heterogeneity; too fine → inflates analytical cost and noise without resolution gain

Sampling Bias Factor (SBF)

0.92–1.08 (i.e., −8% to +8% bias) when core-drill QA/QC is enforced; >±15% indicates critical protocol failure

Quantitative measure of systematic deviation between blasthole assay grade and true in-situ grade, expressed as ratio or % difference

⚡ Engineering Impact:

Uncorrected bias propagates directly into mill feed forecasts and drives erroneous blend ratios and circuit setpoints

Mass Pull Accuracy

±3.5% to ±6.0% for GPS-weighed haul trucks; ±1.2% for calibrated in-pit belt scales

Precision of measured tonnage delivered from a blast round to the crusher or stockpile, relative to modeled volume × density

⚡ Engineering Impact:

Poor mass accuracy invalidates grade-tonnage reconciliation, masking true grade deviation sources

Time Lag (Assay-to-Feed)

48–120 hours for conventional labs; 8–24 hrs for on-site XRF or LIBS systems

Elapsed time between blasthole assay result availability and corresponding material entering the SAG mill feed conveyor

⚡ Engineering Impact:

Longer lags degrade responsiveness of feedback loops, forcing reliance on predictive models instead of real-time adjustment

📐 Key Formulas

Grade Reconciliation Variance (GRV)

GRV = √[Σ(G_assay − G_mill)² / n]

Standard deviation of grade differences between blasthole-assay-weighted feed grade and actual mill feed grade

Typical Ranges:
Porphyry copper (Cu)
0.03–0.09% Cu
Oxide gold (Au)
0.12–0.28 g/t Au
⚠️ GRV ≤ 0.06% Cu or ≤ 0.20 g/t Au for stable circuit operation

Effective Sampling Density (ESD)

ESD = (N_assays × ρ_bulk) / V_blast

Mass-normalized assay frequency, accounting for rock density and blasted volume

Typical Ranges:
Hard porphyry (ρ = 2.7 t/m³)
1.8–3.2 t⁻¹
Weathered cap (ρ = 2.2 t/m³)
1.4–2.5 t⁻¹
⚠️ ESD ≥ 2.0 t⁻¹ for reliable high-grade detection (≥2× cutoff)

🏭 Engineering Example

Escondida Mine, Chile

Copper-Molybdenum Porphyry (quartz monzonite)
Mass Pull Accuracy
±4.2%
Assay Sample Density
2.1 assays/tonne
Compositing Interval
1.0 m
Sampling Bias Factor
0.97
Time Lag (Assay-to-Feed)
72 hrs

🏗️ Applications

  • Open-pit copper porphyry operations
  • Underground massive sulfide (VMS) mines
  • Large-scale gold oxide heap leach facilities

📋 Real Project Case

Open Pit Gold Mine Blast Optimization

Large copper mine expansion in Chile

Challenge: High vibration levels affecting nearby structures
Read full case study →

🎨 Technical Diagrams

Blasthole Assay DataCompositing & Spatial AttributionGrade-Tonnage Reconciliation Engine
AssayMassMill FeedΔG = G_assay − G_millAlert if |ΔG| > 0.10% Cu

📚 References

[1]
CIM Definition Standards – For Mineral Resources and Ore Reserves — Canadian Institute of Mining, Metallurgy and Petroleum (CIM)
[3]
Sampling Theory and Practice for Mining and Metallurgy — International Council on Mining and Metals (ICMM)