Open pit-underground mining options and transitions planning: a mathematical programming framework for optimal resource extraction evaluation

Abstract

Near-surface mineral deposits that extend to great depths are amenable to both open pit mining and/or underground mining. The strategic planning of such mineral deposits often leads to several variations of open pit-underground (OP-UG) mining option(s) and transitions including (a) independent open pit (OP) mining, (b) independent underground (UG) mining, (c) simultaneous open pit and underground (OPUG) mining, (d) sequential OPUG mining, and (e) combinations of simultaneous and sequential OPUG mining. Notable limitations to recent developments in the OP-UG mining options and transitions optimization problem includes one or more of the following: a) lack of rigorous mining optimization approach, b) lack of solution optimality assessment, c) lack of geotechnical consideration for the mining options and transition zones, d) lack of consideration of exhaustive variables for essential UG mining complexes, and e) non-comprehensiveness and inefficiency of the implementation models. The main research objectives are 1) propose an optimization technique for OP-UG mining options and transitions planning, and 2) develop, implement and verify a theoretical optimization framework based on Mixed Integer Linear Programming (MILP) model to determine: a) the most suitable mining option(s) to exploit an orebody; b) the position of the required crown pillar, time and order of development of primary and secondary accesses and main ventilation opening, and the schedule of geotechnical support of the secondary development accesses and stopes if UG mining option is considered; and c) the ore and waste extraction schedules that maximizes the net present value (NPV) of the mining project. MATLAB programming platform was chosen for the MILP formulation implementation and a large-scale optimization solver, IBM ILOG CPLEX, was used for this research. The MILP formulation was tested and implemented with an experimental copper dataset and two real gold deposit case studies. The first case study verified the appropriateness of the optimization technique and strategies used in the MILP framework for open pit-underground mining options and transitions planning. The second and third case studies are implemented with stockpile management and multiple essential underground infrastructures to enhance practicality and rigor of the MILP model. The third case study was additionally evaluated with industry standard software, Whittle, and the results compared to that from the MILP model. The MILP model scheduled the deposit with combined sequential and simultaneous OPUG mining over 8 years mine life while Whittle scheduled the deposit for OP mining over a mine life of 20 years. The NPV generated by the MILP model was $ 4.01 billion while the NPV generated by Whittle Milawa NPV algorithm was $ 2.31 billion, representing about 42.4% loss in financial benefits. The stripping ratio from Whittle OP mining was 2.79 compared to 0.34 from MILP model for the OPUG mining. Analysis of the results showed that, the MILP model significantly avoids the mining of excessive waste to uncover mineralized material by switching from OP to UG mining option. This MILP framework implementation for extraction of deep-seated near-surface deposits demonstrate potential value to a mining project at the prefeasibility stage when the global mining options decisions are guided by a rigorous optimization process. The MILP framework do not evaluate the impact of varying crown pillar dimensions on the mining options.