Calculator D4

Fuel Consumption Modeling for Diesel & Electric Haul Trucks

Fuel consumption modeling predicts how much diesel or electricity a haul truck uses to move material over a specific route and load — like a car’s MPG, but for massive mining trucks.

Typical Scale
A 100-truck fleet consumes 80–120 ML diesel/year; BEV equivalent requires 250–400 GWh/year (≈ 1 medium coal unit)
Key Standards
ISO 8608 (road roughness), SAE J2263 (heavy-duty powertrain testing), IEC 62680 (USB-C for charging interfaces)
Industry Adoption
Rio Tinto (Pilbara), BHP (Olympic Dam), and Vale (S11D) all mandate fuel/energy modeling in mine planning approvals since 2021

⚠️ Why It Matters

1
Inaccurate fuel/energy prediction
2
Over- or under-specification of powertrain and battery capacity
3
Excessive capital expenditure on oversized equipment
4
Reduced payload-cycle efficiency
5
Higher lifecycle GHG emissions and OPEX
6
Non-compliance with decarbonization targets (e.g., SBTi, ICMM Net Zero Roadmap)

📘 Definition

Fuel consumption modeling for diesel and electric haul trucks is a physics- and data-driven engineering methodology that quantifies energy demand per ton-kilometer under dynamic operational conditions, accounting for vehicle mass, grade, rolling resistance, aerodynamic drag, drive-train efficiency, battery state-of-charge (for electric), and duty cycle. It integrates empirical powertrain calibration, terrain digital elevation models (DEMs), and real-time telematics to support fleet optimization, infrastructure planning, and emissions forecasting in surface and underground mine transport systems.

🎨 Concept Diagram

Load PointDump PointGrade +7.3%Rolling ResistanceDiesel EngineBattery + Motor

AI-generated illustration for visual understanding

💡 Engineering Insight

Never trust a fuel model built solely on manufacturer ‘rated’ power curves — real-world diesel engines operate 15–25% below peak efficiency 70% of the time due to transient loading, ambient temperature, and aftertreatment backpressure. For battery-electric trucks, the single largest source of model error is inaccurate SOC-to-energy mapping under high C-rate discharge (>1.5C) and sub-zero temperatures; always validate using calorimetrically referenced cell-level test data, not just pack voltage.

📖 Detailed Explanation

At its core, fuel consumption modeling starts with Newtonian mechanics: the sum of forces acting on the truck — gravity (grade), inertia (acceleration), rolling resistance, and aerodynamic drag — determines required tractive power. This power, multiplied by time, yields energy demand. For diesel trucks, that energy maps to fuel mass via lower heating value (LHV ≈ 42.5 MJ/kg) and engine efficiency; for electric trucks, it maps to kWh drawn from the battery, adjusted for inverter, motor, and thermal losses.

Going deeper, modern models incorporate time-varying dynamics: gear-shift timing affects engine operating point and thus BSFC (brake-specific fuel consumption); tire slip ratio modifies effective rolling resistance; and battery internal resistance rises with temperature and SOC, causing voltage sag and reduced usable capacity. These nonlinearities require either high-fidelity co-simulation (e.g., AVL CRUISE™ + MATLAB/Simulink) or physics-informed machine learning (PIML) trained on multi-year telematics archives.

At the advanced level, models must address fleet-scale interactions: traffic density alters acceleration/deceleration patterns; battery degradation changes regen capability and charging time windows; and ore variability (e.g., wet vs dry, density shifts) modifies payload mass and thus GVW distribution. The most robust implementations embed uncertainty quantification (e.g., Monte Carlo on RRC and grade inputs) and link directly to mine planning software (e.g., Deswik, MineSuite) to evaluate trade-offs between haul road maintenance cost, energy cost, and production schedule adherence.

🔄 Engineering Workflow

Step 1
Step 1: Define haul cycle geometry (origin, destination, elevation profile, road surface class)
Step 2
Step 2: Characterize truck powertrain (engine map / motor torque-speed curve, transmission ratios, brake types)
Step 3
Step 3: Acquire real-world telemetry (speed, throttle/brake position, SOC, GVW, GPS) across representative shifts
Step 4
Step 4: Calibrate physics-based model (e.g., ISO 8608-inspired road excitation + SAE J2263 powertrain loss model) using regression or ML-augmented parameter estimation
Step 5
Step 5: Validate against field fuel/energy metering (e.g., Coriolis flow meters, DC shunt + BMS telemetry) over ≥100 haul cycles
Step 6
Step 6: Integrate into fleet dispatch system for real-time energy-aware scheduling and battery state forecasting
Step 7
Step 7: Update model quarterly using degradation-adjusted parameters (e.g., tire wear, battery EOL capacity loss)

📋 Decision Guide

Rock/Field Condition Recommended Design Action
High-Grade Haul Route (>8% ascending, >6% descending) with frequent stops Deploy battery-electric trucks with high-torque low-RPM motors and ≥40% regen recovery; optimize haul cycle sequencing to maximize downhill regeneration
Low-Grade, Long-Haul (>3 km round trip), Dry Gravel Surface (RRC ≈ 0.015) Use high-efficiency diesel-electric drives (e.g., AC traction) with predictive cruise control; avoid oversizing battery capacity if electrifying
Underground Ramp System (8–10% grade, confined ventilation, limited charging infrastructure) Select trolley-assisted or opportunity-charged battery-electric trucks with thermal management; model battery SOH degradation using real-world charge/discharge profiles
Mixed Fleet Operation (Diesel + Battery-Electric) on Shared Haul Roads Implement dynamic traffic management (DTM) to separate high-braking-demand descents for electric units; calibrate fuel/energy models using shared GNSS-telematics datasets

📊 Key Properties & Parameters

Gross Vehicle Weight (GVW)

210–400 tonnes (off-highway rigid frame haul trucks)

Total mass of the fully loaded truck including chassis, engine/battery, fuel/electrolyte, operator, and payload.

⚡ Engineering Impact:

Dominates tractive effort requirement and regenerative braking potential; directly scales rolling resistance and grade-related power demand.

Road Grade Profile

−8% to +12% (common in open-pit mines; ±0.14 rad)

Vertical slope (as % or radians) along the haul cycle, derived from high-resolution DEMs or survey-grade GNSS.

⚡ Engineering Impact:

Grades >5% increase diesel fuel use by 20–40% per km; governs optimal gear selection and electric motor thermal derating.

Rolling Resistance Coefficient (RRC)

0.012–0.035 (gravel, well-maintained haul roads)

Dimensionless factor representing energy loss due to tire deformation and road surface interaction, normalized per unit weight.

⚡ Engineering Impact:

A 0.005 increase in RRC raises fuel consumption by ~7–9% at constant speed; highly sensitive to road moisture, aggregate size, and compaction.

Drive-Train Efficiency (η)

0.32–0.38 (diesel-mechanical), 0.82–0.89 (battery-electric drivetrain, including inverter & motor losses)

Ratio of mechanical power delivered to wheels versus chemical (diesel) or electrical (battery) input power.

⚡ Engineering Impact:

Lower η forces higher fuel/energy throughput to achieve same work — critical for TCO analysis and battery sizing.

Regenerative Braking Recovery Rate

25–45% (modern battery-electric haul trucks; <5% for diesel-hydraulic retarders)

Fraction of kinetic + potential energy recovered during descent and braking, stored back in battery or dissipated.

⚡ Engineering Impact:

Each 10% improvement reduces net energy consumption by 3–6% on mixed-grade cycles — decisive for ROI on electrification.

📐 Key Formulas

Tractive Power Requirement

P_trac = (m·g·sinθ + m·a + C_r·m·g·cosθ + 0.5·ρ·C_d·A·v²) · v

Instantaneous mechanical power needed at wheels to overcome grade, acceleration, rolling resistance, and aerodynamic drag.

Variables:
Symbol Name Unit Description
P_trac Tractive Power W Instantaneous mechanical power needed at the wheels
m Vehicle Mass kg Total mass of the vehicle
g Gravitational Acceleration m/s² Standard acceleration due to gravity
θ Road Grade Angle rad Angle of incline (positive uphill)
a Vehicle Acceleration m/s² Longitudinal acceleration of the vehicle
C_r Rolling Resistance Coefficient dimensionless Coefficient representing resistance due to tire deformation and road interaction
ρ Air Density kg/m³ Mass density of ambient air
C_d Drag Coefficient dimensionless Dimensionless quantity quantifying aerodynamic drag
A Frontal Area Projected cross-sectional area of the vehicle perpendicular to motion
v Vehicle Speed m/s Instantaneous forward speed of the vehicle
Typical Ranges:
Loaded ascent (10% grade)
1,400–2,100 kW
Empty descent (8% grade, regen active)
−800 to −1,300 kW
⚠️ Motor/generator continuous rating must exceed 110% of max positive P_trac and absorb 105% of max negative P_trac without thermal shutdown.

Diesel Fuel Consumption (Volumetric)

Ḟ_fuel = (P_e · BSFC) / (ρ_fuel · η_trans)

Volumetric fuel flow rate based on engine brake power, brake-specific fuel consumption, fuel density, and transmission efficiency.

Variables:
Symbol Name Unit Description
Ḟ_fuel Volumetric fuel flow rate m³/s Volume of diesel fuel consumed per unit time
P_e Engine brake power W Effective mechanical power output of the engine
BSFC Brake-specific fuel consumption kg/(kW·h) or kg/J Mass of fuel consumed per unit energy output of the engine
ρ_fuel Fuel density kg/m³ Mass per unit volume of diesel fuel
η_trans Transmission efficiency dimensionless Ratio of output power to input power of the transmission system
Typical Ranges:
Rated load, 25°C ambient
110–145 L/h
Part-load, high-altitude (>3,000 m)
125–165 L/h (due to reduced η_engine & η_trans)
⚠️ BSFC > 220 g/kWh indicates need for engine diagnostics or aftertreatment service.

Battery Energy Consumption (Net)

E_net = ∫(P_batt(t) dt) = ∫((P_trac(t)/η_mot·inv) + P_aux(t)) dt − E_regen

Net battery energy used per haul cycle, accounting for drivetrain losses, auxiliaries, and regenerative recovery.

Variables:
Symbol Name Unit Description
E_net Net Battery Energy Consumption Joules (J) or kWh Net energy drawn from the battery per haul cycle, accounting for traction power losses, auxiliary loads, and subtracting regenerated energy
P_batt(t) Battery Power Watts (W) Instantaneous power delivered by the battery at time t
P_trac(t) Traction Power Watts (W) Instantaneous mechanical power delivered to the wheels at time t
η_mot·inv Motor-Inverter Efficiency dimensionless Combined efficiency of motor and inverter converting battery electrical power to mechanical traction power
P_aux(t) Auxiliary Power Watts (W) Instantaneous power consumed by auxiliary systems (e.g., cooling, lighting, HVAC) at time t
E_regen Regenerated Energy Joules (J) or kWh Total energy recovered and returned to the battery during braking/regeneration per haul cycle
Typical Ranges:
Medium-grade cycle (4.5% avg), 320 t GVW
85–105 kWh/haul
Steep-cycle (8.2% avg), cold start (<5°C)
120–145 kWh/haul
⚠️ Sustained discharge >1.2C degrades NMC battery life by >35% per 1,000 cycles; limit to ≤0.9C average unless thermally managed.

🏭 Engineering Example

Chuquicamata Open Pit, Codelco, Chile

Porphyry copper ore (in situ density 2.65 g/cm³; blasted muck density 1.82 g/cm³)
GVW
345 tonnes
RRC
0.024
Max_Grade
+10.2%
Avg_Haul_Distance
2.8 km (one-way)
Diesel_Fuel_Use_Rate
128 L/h (loaded ascent, 25 km/h avg)
Battery_Energy_Use_Rate
112 kWh/haul (Cat 794 AC, validated telemetry, 2023 Q3)

🏗️ Applications

  • Mine fleet decarbonization roadmap development
  • Haul road design optimization (grade smoothing, camber, surfacing)
  • Battery procurement & charging infrastructure sizing
  • Real-time dispatch optimization (energy-aware routing)

📋 Real Project Case

Chilean Copper Mine: Autonomous Haul Fleet Deployment

A Tier-1 copper mine in the Atacama Desert, northern Chile, deployed an autonomous haul fleet across its open-pit operation. The site processes ~450 ktpd of ore and waste, with a 2.8-km average haul distance and 320-m vertical lift. The project involved retrofitting and integrating 42 autonomous 290-tonne CAT 794 AC electric drive haul trucks into existing dispatch and traffic management systems.

Challenge: Achieving safe, reliable, and productive autonomous haulage under extreme environmental conditions (...
Chilean Copper Mine: Autonomous Haul Fleet DeploymentDTDigital TwinSFSensor FusionECEdge ComputePCPhased Commissioningd = 187.3 mBraking distanceA = 22.6 dBLiDAR attenuationσ_pos = 0.17 mGNSS-RTK (3D RMS)Extreme EnvironmentAltitude: 3200 m ASL • Temp: −5°C to 42°C • Dust: ρ = 1200 μg/m³ • Steep/winding roads
Read full case study →

Frequently Asked Questions

What distinguishes fuel consumption modeling for haul trucks from standard vehicle fuel economy estimates?
Unlike standard MPG or kWh/100 km estimates, haul truck fuel consumption modeling is operationally contextual and physics-based — it dynamically accounts for payload weight, terrain grade (via DEMs), rolling resistance, aerodynamic drag, drivetrain efficiency, battery state-of-charge (for electric variants), and real-world duty cycles (e.g., loading time, idle duration, acceleration profiles). This enables accurate per-ton-kilometer energy demand quantification specific to mine transport environments.
Why is terrain modeling — particularly Digital Elevation Models (DEMs) — critical in this methodology?
DEM data provides precise elevation profiles along haul routes, enabling accurate calculation of gravitational work (i.e., energy required to climb grades) and regenerative braking potential (on descents) — especially vital for electric trucks. Without high-resolution terrain input, models cannot reliably simulate the dominant energy sinks and gains inherent in off-highway mining operations.
How does the model handle differences between diesel and electric powertrains?
For diesel trucks, the model uses empirically calibrated engine maps (fuel rate vs. torque/speed) and transmission efficiency curves. For electric trucks, it integrates motor/inverter efficiency maps, battery voltage-current-SoC relationships, thermal derating effects, and regenerative braking recovery rates. Both pathways converge on net energy demand per ton-km but reflect fundamentally different loss mechanisms and operational constraints.
Can this modeling approach be applied to both surface and underground mining operations?
Yes — the framework is adaptable to both environments. Surface applications leverage GPS-georeferenced DEMs and ambient conditions; underground applications substitute laser-scanned tunnel profiles and ventilation-constrained thermal limits, while adjusting rolling resistance (due to rubber-on-rock vs. asphalt) and aerodynamic assumptions. Telematics integration remains central in both cases.
What data inputs are required to build and validate a reliable fuel consumption model?
Essential inputs include: (1) vehicle specifications (mass, axle configuration, tire type, powertrain architecture), (2) route geometry (DEM or 3D scan data), (3) duty cycle telemetry (speed, torque, brake usage, SoC, engine/fuel rate logs), (4) environmental data (temperature, altitude, humidity), and (5) empirical powertrain calibration data. Validation requires synchronized field measurements across representative loads, grades, and operating modes.

🎨 Technical Diagrams

OriginDestination+6.2%+10.4%
DieselBEVTrolley
Regen Recovery: 38%Battery Discharge: 67%Auxiliary Loads: 100%

📚 References