🎓 Lesson 15
D5
Multi-Objective Tradeoff Analysis: Cost, Throughput, and Carbon
Multi-objective tradeoff analysis is a way to find the best balance between competing goals—like spending less money, moving more material, and emitting less carbon—when designing a mining or blasting operation.
🎯 Learning Objectives
- ✓ Calculate Pareto-efficient blast designs by varying burden, spacing, and powder factor while tracking cost, tonnes blasted per shift, and CO₂e per tonne
- ✓ Analyze tradeoff surfaces using normalized objective plots to identify inflection points where marginal carbon savings begin to degrade throughput
- ✓ Design a constrained multi-objective optimization model in Python or Excel for a given open-pit bench scenario with at least three decision variables and three objectives
- ✓ Explain how carbon intensity metrics (e.g., kg CO₂e/tonne ROM) interact with logistics chain bottlenecks (e.g., shovel-truck cycle time, crusher throughput)
- ✓ Apply weighting schemes and preference articulation methods (e.g., TOPSIS) to rank alternative logistics configurations under stakeholder-defined priorities
📖 Why This Matters
Mining operations face unprecedented pressure: investors demand cost discipline, regulators enforce Scope 1 & 2 emissions limits, and communities expect sustainable productivity. A 5% reduction in drilling cost may increase fragmentation variability—raising crushing energy use and carbon footprint. A 10% larger truck fleet boosts throughput but raises diesel consumption and maintenance costs. Multi-objective tradeoff analysis equips engineers to move beyond 'best on one metric' to 'best balanced across all critical metrics'—a skill now required in ESG-aligned mine planning and reporting (e.g., ICMM Climate Principles, GRI 305).
📘 Core Principles
Tradeoff analysis begins with formalizing objectives as mathematical functions: Total Cost (C), Throughput (T), and Carbon Intensity (Γ). These are typically non-commensurable (different units, scales, and directional preferences)—so they must be normalized and constrained. The Pareto frontier defines solutions where no objective can improve without worsening at least one other. Key theoretical constructs include: (1) Objective space mapping (C–T–Γ), (2) Decision variable sensitivity (e.g., how burden affects rock size distribution → crusher energy → Γ), (3) Constraint hierarchies (geotechnical stability > regulatory emissions cap > budget ceiling), and (4) Preference incorporation via scalarization (weighted sum, ε-constraint, or achievement functions). Real-world implementation requires integrating empirical models (e.g., Kuz-Ram fragmentation, Caterpillar fuel maps, IPCC Tier 1 emission factors) into optimization frameworks.
📐 Normalized Weighted Objective Function
This scalarized function converts multi-objective optimization into a tractable single-objective problem while preserving relative importance. Weights reflect strategic priorities (e.g., carbon weight = 0.5 if net-zero commitment is binding). Normalization prevents unit dominance (e.g., $1M cost vs. 0.002 kg CO₂e/tonne).
Weighted Sum Scalarization (Normalized)
S = w_C · nC + w_T · nT + w_Γ · nΓConverts multi-objective optimization into a single scalar objective for ranking or solver input, using normalized and weighted objectives.
Variables:
| Symbol | Name | Unit | Description |
|---|---|---|---|
| S | Scalarized Score | dimensionless | Composite performance index (higher = better, per defined weights) |
| w_C | Cost Weight | dimensionless | Relative importance assigned to cost minimization (0 ≤ w_C ≤ 1, Σw = 1) |
| nC | Normalized Cost | dimensionless | (C_max − C_i)/(C_max − C_min); ranges [0,1] where 1 = best cost performance |
| w_T | Throughput Weight | dimensionless | Relative importance assigned to throughput maximization |
| nT | Normalized Throughput | dimensionless | (T_i − T_min)/(T_max − T_min); 1 = highest throughput |
| w_Γ | Carbon Weight | dimensionless | Relative importance assigned to carbon intensity minimization |
| nΓ | Normalized Carbon Intensity | dimensionless | (Γ_max − Γ_i)/(Γ_max − Γ_min); 1 = lowest carbon intensity |
Typical Ranges:
ESG-focused brownfield expansion: w_C = 0.3–0.5, w_T = 0.2–0.4, w_Γ = 0.3–0.5
Cost-constrained greenfield startup: w_C = 0.5–0.7, w_T = 0.2–0.3, w_Γ = 0.1–0.2
💡 Worked Example
Problem: A copper mine evaluates three blast designs (A, B, C) for a 15-m bench. Measured outcomes: Design A → Cost = $125,000, Throughput = 18,500 t/shift, Γ = 0.82 kg CO₂e/t. Design B → $138,000, 21,300 t/shift, 0.97 kg CO₂e/t. Design C → $149,000, 22,600 t/shift, 1.14 kg CO₂e/t. Management assigns weights: w_C = 0.4, w_T = 0.3, w_Γ = 0.3. Normalize each objective to [0,1] using min/max across designs (min cost = $125k, max throughput = 22,600 t, min Γ = 0.82). Calculate weighted score for each design.
1.
Step 1: Normalize Cost (minimize): nC = (C_max − C_i)/(C_max − C_min) → nC_A = (149−125)/(149−125) = 1.0; nC_B = (149−138)/24 = 0.46; nC_C = 0.0
2.
Step 2: Normalize Throughput (maximize): nT = (T_i − T_min)/(T_max − T_min) → T_min = 18,500 → nT_A = 0.0; nT_B = (21,300−18,500)/4,100 = 0.68; nT_C = 1.0
3.
Step 3: Normalize Carbon (minimize): nΓ = (Γ_max − Γ_i)/(Γ_max − Γ_min) → Γ_max = 1.14 → nΓ_A = (1.14−0.82)/0.32 = 1.0; nΓ_B = (1.14−0.97)/0.32 = 0.53; nΓ_C = 0.0
4.
Step 4: Compute weighted score S_i = w_C·nC_i + w_T·nT_i + w_Γ·nΓ_i → S_A = 0.4(1.0)+0.3(0.0)+0.3(1.0) = 0.70; S_B = 0.4(0.46)+0.3(0.68)+0.3(0.53) = 0.54; S_C = 0.4(0.0)+0.3(1.0)+0.3(0.0) = 0.30
Answer:
Design A scores highest (0.70), reflecting strongest balance under stated weights—despite lowest throughput and highest carbon, its cost and carbon advantages dominate. This illustrates how weighting drives tradeoff resolution.
🏗️ Real-World Application
At Newmont’s Boddington Mine (Western Australia), engineers applied multi-objective tradeoff analysis to redesign the primary blast pattern for the South Pit. Using Deswik Blast and @RISK, they modeled 216 scenarios varying burden (5.2–6.8 m), spacing (6.0–7.5 m), and explosive type (ANFO vs. emulsion). Objectives: minimize total blast cost ($/t), maximize muck pile uniformity (measured by Rosin-Rammler slope n ≥ 1.4), and minimize Scope 1 CO₂e (via emulsion’s lower NOₓ and higher energy density). The Pareto frontier revealed that increasing burden beyond 6.1 m degraded n below 1.35—triggering secondary breaking and raising downstream diesel use. The selected design (burden = 6.1 m, spacing = 6.7 m, 70% emulsion) reduced CO₂e by 8.3% and cost by 4.1% versus baseline, with no throughput loss—validated over 12 months of production data.
🔧 Interactive Calculator
🔧 Open Mine Logistics Chain Optimization Calculator📋 Case Connection
📋 Chilean Iron Ore Export Corridor Optimization
Chronic rail delays causing port demurrage penalties and stockpile overflow
📋 South African Platinum Group Metals Stockpile Optimization
Overstocking of lower-grade material due to inflexible blending schedules and forecast errors