501 3.3. Results For the implementation, four consecutive periods were evaluated to represent the regular activities over two days (each period represents a 10-hr shift). The solver selected for the task was Gurobi 12.0.0 executed in AMPL software. As mentioned before, when categories are fixed based on a break-even cutoff grade in either ore or waste, marginal material is labeled as waste and the corresponding grade is replaced by zero metal content. When this happens, the model is forced to only consider ore-labeled muck piles for possible allocations to satisfy processing facility requirements, even if tonnages are insufficient or grades are away from the target objectives. Table reports the summary results for this implementation over the seven scenarios proposed. From P5 to P70, the results are similar in performance and are only differentiable by the increments generated by the metal content increase and small variances in the truck allocations (P50 highlighted in yellow represents the average case). This is because there is no significant change in the muck piles’ categorization for the shovels to be dynamically allocated, where there is always a pre-requisite to strip at least one of the wastecategorized muck piles to have access to enough ore to feed the mill. In contrast, P90 and P95 (best case scenarios) break the uniformity by proposing a plan where waste stripping is not needed immediately to satisfy processing facility requirements. Therefore, truck fleet use is reduced, penalties in ore production are dismissed, and total revenue is substantially increased. Table 2 – Summary results for fixed categories approach. KPIs P5 Scenario1 P10 Scenario2 P30 Scenario3 P50 Scenario4 P70 Scenario5 P90 Scenario6 P95 Scenario7 Ore Production (tons) 80,000 80,000 80,000 80,000 80,000 100,000 100,000 Waste Stripping (tons) 30,000 30,000 30,000 30,000 30,000 0 0 Average Grade (%Cu) 0.3491% 0.3514% 0.3564% 0.3600% 0.3635% 0.3289% 0.3313% Recovered Metal (ton-Cu) 225.13 226.94 230.78 233.60 236.28 263.08 265.04 Number of Active Trucks 61 61 61 61 63 66 65 150-ton Truck Trips 568 568 568 572 544 472 484 200-ton Truck Trips 124 124 124 121 142 146 137 Total Truck Trips 692 692 692 693 686 618 621 Gross Revenue (USD) 2,071,178 2,087,885 2,123,139 2,149,120 2,173,776 2,420,372 2,438,332 Shovel Cost 7,579 7,580 7,580 7,580 7,579 7,255 7,257 Truck Cost 167,859 167,890 167,952 167,844 168,158 171,474 171,298 Processing Cost 1,600,000 1,600,000 1,600,000 1,600,000 1,600,000 2,000,000 2,000,000 Penalties on Production 200,000 200,000 200,000 200,000 200,000 0 0 Penalties on Grades 11,665 13,422 17,124 19,948 23,382 22,159 17,303 Revenue before Penalties 295,741 312,415 347,607 373,697 398,039 241,643 259,777 Total Revenue (USD) 84,075 98,992 130,483 153,749 174,656 219,484 242,474 Following the summary results, the analysis of the most important performance indicators can clarify certain results. Values such as average grade at the processing facility, recovered metal, cumulative truck trips, and cumulated revenue are significant for the interpretation of the sensitivity of the short-term planning to geologic risk (accuracy on the block grades). In Figure , this cumulative variance can be observed over time, clearly displaying the impact of not having timed access to desired grade material at the correct moment (period 2). Although most scenarios have a total revenue increment towards the end of the time horizon, these drops are a major concern in the context of the processing plant’s regular operations, due to
RkJQdWJsaXNoZXIy MTM0Mzk2