🎓 Lesson 8
D4
Validating SMVs with Statistical Process Control
SMV validation with Statistical Process Control means using data charts and statistics to check if the time it takes workers to do a mining or blasting task is consistent, fair, and improving over time.
🎯 Learning Objectives
- ✓ Calculate process capability indices (Cp and Cpk) for SMV-based cycle times using sample data
- ✓ Design and interpret X-bar and R control charts for SMV validation in drilling or mucking operations
- ✓ Analyze SMV stability using Western Electric rule violations and identify assignable causes
- ✓ Apply ANOVA to compare SMV consistency across multiple blast crews or shift rotations
- ✓ Explain how SPC-validated SMVs support equitable incentive pay and regulatory compliance (e.g., MSHA Part 46 recordkeeping)
📖 Why This Matters
In underground and open-pit mining, SMVs drive labor costing, crew scheduling, productivity bonuses, and safety compliance—but an unvalidated SMV is just an educated guess. When SMVs drift due to fatigue, tool wear, or undocumented method changes, they erode trust, inflate costs, and mask real inefficiencies. SPC transforms SMV engineering from static snapshots into a living, auditable system—ensuring fairness, enabling root-cause improvement, and meeting ISO 9001 and MSHA audit requirements for documented process control.
📘 Core Principles
SMV validation via SPC rests on three pillars: (1) Data integrity—cycle times must be collected under standardized conditions (same equipment, operator qualification level, rock mass rating, and environmental controls); (2) Process stability—time data must exhibit only common-cause variation (i.e., be 'in control' per Shewhart criteria); and (3) Capability—measured times must fit within specification limits (e.g., ±10% of target SMV) with adequate margin (Cpk ≥ 1.33). SPC distinguishes between natural variation (expected noise) and special-cause variation (e.g., drill bit failure, misaligned burden spacing), which must be eliminated before SMV acceptance. Control chart rules (e.g., 2-of-3 points beyond 2σ) signal when revalidation is mandatory—not merely recalibration.
📐 Process Capability Index (Cpk)
Cpk quantifies how well the SMV-based process fits within its tolerance band, accounting for both spread and centering. A Cpk < 1.0 indicates unacceptable variation; ≥1.33 is industry-accepted for critical labor standards. Used after confirming statistical control (via control charts).
Cpk (Process Capability Index)
Cpk = min[(USL − μ)/(3σ), (μ − LSL)/(3σ)]Measures process capability relative to nearest specification limit, penalizing off-centering.
Variables:
| Symbol | Name | Unit | Description |
|---|---|---|---|
| USL | Upper Specification Limit | minutes | Maximum allowable time per unit (e.g., SMV × 1.10) |
| LSL | Lower Specification Limit | minutes | Minimum allowable time per unit (e.g., SMV × 0.90) |
| μ | Process Mean | minutes | Average observed cycle time from stable process data |
| σ | Process Standard Deviation | minutes | Within-subgroup estimate of variation (e.g., R̄/d₂ from R-chart) |
Typical Ranges:
Validated SMV for drilling: Cpk ≥ 1.33
Marginally acceptable SMV: Cpk = 1.00–1.32
Unvalidated SMV requiring intervention: Cpk < 1.00
💡 Worked Example
Problem: A mucking operation has an SMV target of 4.2 min/ton. Field data from 25 consecutive 1-ton cycles shows mean = 4.38 min, standard deviation = 0.21 min. Specification limits are USL = 4.62 min (target +10%), LSL = 3.78 min (target −10%). Calculate Cpk.
1.
Step 1: Compute distance from mean to nearest spec limit: min(4.62 − 4.38, 4.38 − 3.78) = min(0.24, 0.60) = 0.24
2.
Step 2: Divide by 3 × σ: 0.24 / (3 × 0.21) = 0.24 / 0.63 ≈ 0.381
3.
Step 3: Interpret: Cpk = 0.38 < 1.0 → process is incapable; SMV cannot be validated without corrective action (e.g., bucket calibration, operator retraining).
Answer:
Cpk = 0.38, indicating severe process incapability—SMV requires revision and root-cause investigation before deployment.
🏗️ Real-World Application
At Newmont’s Boddington Mine (Western Australia), SMVs for blast-hole drilling were invalidated after SPC revealed 7 consecutive points above the X-bar chart centerline—traced to progressive hydraulic pump degradation in drill rigs. After implementing predictive maintenance triggers at R-chart range thresholds, Cpk improved from 0.82 to 1.41 within 3 weeks. This enabled accurate labor cost allocation across 12 shift teams and resolved disputes over performance-based pay—documented in their 2022 Operational Excellence Report (Newmont Technical Bulletin No. 22-07).
📋 Case Connection
📋 Food Processing Packaging Line Throughput Lift
Peak demand periods caused 40% throughput shortfall; reliance on temporary staff with inconsistent training