ABSTRACT
Mine production schedules are commonly based on block model estimates which only provide a single expected value realisation of deposit characteristics, and cannot account for the true variability of the deposit’s characteristics. Production schedules based on a weighted average estimator cannot predict the true deviations in production targets, and their use will inevitably result in mine-mill reconciliation errors. Simulation reproduces the variability of the deposit conditional on the samples. Given sufficient simulated realisations of the deposit, the uncertainty associated with production variables can be quantified in each block. Stochastic Programming (SP) methods, specifically simple recourse models, are examined as a means of extending Linear- and Integer Programming-based production scheduling algorithms into a stochastic optimisation paradigm. In the SP approach to scheduling, the distribution of the production variable in each block is used as input to a single optimisation which will determine a sequence of block extraction which accounts for deposit uncertainty. A Simple Recourse model is presented to demonstrate the value of the stochastic solution over the expected value solution.
