Reduce Energy Consumption

Surface coal mining operations rely on truck-and-shovel systems to remove overburden — a fuel-intensive process with significant energy waste embedded in poorly understood cycle times, payload variability, and equipment load factors. Without a quantitative baseline, operations teams had no systematic way to rank energy-saving opportunities or predict the impact of operational changes before implementing them. The cost of the status quo was measurable: inefficient shovel load factors, suboptimal truck dispatch, and overall fuel efficiency of just 19.09 tons per gallon across the combined truck-and-shovel operation — leaving substantial reduction potential untapped.

The research team used Arena, Rockwell Automation's discrete-event simulation platform, to build a stochastic process simulation model of the truck-and-shovel overburden removal operation. Energy audits were conducted on-site to collect empirical cycle time and payload data. The chi-squared goodness-of-fit test was applied to fit theoretical probability distributions to this data, which were then used to parameterize the stochastic processes within the Arena digital twin. The validated model — benchmarked against actual truck fuel consumption measurements — enabled the team to run controlled experiments across a range of operational strategies without disrupting live production, producing a ranked list of high-impact energy-saving improvement options.

Implementing the top-ranked operational strategies identified through simulation produced a near 49% reduction in energy consumption across the operation. Baseline measurements from Mine 1 established the pre-optimization state:

• Shovel engine average load factor: 66.78%, corresponding to a fuel consumption rate of 35.36 gals/hr
• Shovel fuel efficiency: 39.29 tons/gal
• Average truck fuel consumption: 3.68 gals/cycle (efficiency: 37.14 tons/gal)
• Combined truck-and-shovel fuel efficiency: 19.09 tons/gal

The simulation-driven approach allowed teams to identify and prioritize interventions that delivered the largest energy impact before any operational changes were committed.

• Simulation before intervention: Building a validated digital twin before changing operations avoids costly trial-and-error and enables risk-free comparison of strategies at scale.
• Data quality is foundational: Accurate distribution fitting (validated with goodness-of-fit tests) is what separates a reliable model from an optimistic one — invest in rigorous data collection upfront.
• Rank, don't just identify: Producing a prioritized list of improvement options — not just a binary go/no-go — lets operations teams sequence changes by expected impact.
• Validate against real consumption data: Benchmarking the model against measured fuel consumption before using it for decisions is non-negotiable for credibility with operations leadership.