ToolFusion
📋 Case Study

Mine Safety & Risk Management Case Study 2

Commercial application

🏗️ Project Overview

A Tier-1 iron ore mining operation in the Pilbara region of Western Australia, producing 85 Mtpa (million tonnes per annum) across three open-pit mines and a centralized processing hub. The site employs 2,400 personnel across shifts and operates under strict regulatory oversight from the WA Department of Mines, Industry Regulation and Safety (DMIRS).

🎯 Challenge

High-frequency occurrence of rockfall events on active pit slopes (particularly in the newly developed North Pit), resulting in near-miss incidents, unplanned production stoppages averaging 3.2 hours/week, and elevated risk scores in the site’s ALARP (As Low As Reasonably Practicable) register for slope stability and personnel exposure.

🔧 Design Approach

Adopted an integrated geotechnical–human factors–digital twin risk management framework: (1) High-resolution LiDAR and photogrammetric slope monitoring combined with real-time displacement analytics; (2) Probabilistic slope stability modeling using Monte Carlo simulation calibrated to site-specific rock mass properties (RMR, GSI); (3) Dynamic exclusion zone automation triggered by displacement thresholds; (4) Integration with the mine’s existing JSA (Job Safety Analysis) and permit-to-work systems via API-enabled digital twin platform.

📐 Design Diagram

Integrated Geotechnical–Human Factors–Digital Twin FrameworkRockfall14.2% PoFDigital TwinAPI-enabledReal-time LiDAR + Displacement AnalyticsMonte Carlo Slope Stability ModelDynamic Exclusion Zone (R = 42.6 m)ALARP CBR = 0.68(Cost-Benefit Ratio < 1 → ALARP achieved)JSA & Permit-to-Work Integration

AI-generated project design illustration

📐 Key Calculations

Probability of Failure (PoF) for North Pit West Wall

PoF = ∫ P(FS < 1.0) · f(FS) dFS, where FS is factor of safety sampled from lognormal distribution (μ=1.32, σ=0.18) derived from 127 borehole RQD/GSI datasets
Result: 0.142 (14.2%)
Quantified unacceptable risk level exceeding DMIRS target of <5% PoF for critical slopes, justifying immediate engineering intervention.

Dynamic Exclusion Zone Radius

R = k × √(D_max × H), where D_max = max measured displacement rate (mm/day), H = wall height (m), k = empirical coefficient (0.85 based on historical rockfall trajectory analysis)
Result: 42.6 m
Enabled automated real-time redefinition of safe working boundaries, reducing manual survey-dependent hazard zoning delays by 97%.

ALARP Cost-Benefit Ratio (CBR)

CBR = (Annualised Risk Reduction Cost) / (Expected Annual Loss Reduction), where loss includes downtime, injury cost, and regulatory penalty exposure
Result: 0.68
CBR < 1.0 confirmed that proposed mitigation (instrumented slope reinforcement + AI-driven alerting) was economically justified and met ALARP criteria.

📊 Results

Metrics: Rockfall incidents reduced by 89% (from 23 to 2.5/year), Unplanned stoppages decreased to 0.3 hours/week, Slope-related LTI (Lost Time Injury) frequency rate reduced from 2.1 to 0.0, Regulatory non-conformance findings dropped from 11 to 1 per audit cycle
Implementation of the integrated risk management system achieved full compliance with DMIRS Code of Practice for Mine Slope Management within 11 months, eliminated high-consequence slope failure pathways, and delivered A$14.2M in cumulative operational savings over Year 1 (primarily from avoided downtime and insurance premium reduction).

💡 Lessons Learned

  • •Geotechnical models must be continuously updated with real-time monitoring data—not treated as static design inputs
  • •Operator trust in automated exclusion zones required co-design with frontline crews and transparent algorithm validation

✅ Key Takeaways

  • 1Risk-informed engineering decisions require coupling probabilistic geomechanical modeling with operational human-system integration—not siloed technical solutions