2071 6. VALIDATION OF PRODUCTION PLANS 6.1 From Point Estimation to Probability Management Production plans in underground mining are commitments that operations make to their organizations and shareholders. The way these commitments are developed has direct implications for their fulfilment: a plan built on point estimates of nominal capacity without explicit consideration of operational variability is a plan whose actual probability of compliance is unknown and, frequently, significantly below 100%. Simulation transforms this situation by allowing production plans to be developed — and evaluated — as probabilistic commitments. Instead of setting a daily target of X tonnes assuming the operation will function at average levels, the planner can set a target that the model indicates will be achieved with a probability of 80% or 90%, depending on the desired confidence level. This way of planning is more honest, more informative, and useful for operational management (Figure 5). Figure 5 - Simulated daily production distribution versus deterministic plan estimate (generic illustrative scenario). The nominal plan value (dashed line) represents a single point estimate that ignores operational variability. The simulation delivers a full probability distribution, revealing the frequency with which actual production falls below the plan target and enabling planners to set production commitments at explicit confidence levels (e.g., P80 or P90). 6.2 Production Plan Sensitivity Analysis Probabilistic validation of the plan is complemented by a sensitivity analysis that identifies the factors most influencing the probability of compliance. This analysis answers questions such as: how much does the probability of meeting the plan fall if MHS availability decreases by 5 percentage points? How much does it increase if an additional LHD is added to the most productive sector? What happens to compliance if the hang-up frequency in the next quarter corresponds to the 90th percentile of historical difficulty? The answers to these questions, obtained by running the model under different parameter combinations, generate a plan sensitivity map that enables management to identify the most relevant risk factors and design specific contingency protocols for each. This map converts uncertainty — which in traditional methods is simply ignored — into actionable information for management.
RkJQdWJsaXNoZXIy MTM0Mzk2