Simulation and Optimization in Open Pit Mines

"Increasing demand and market economics have pushed the mining industry to adopt improved mining practices and minimize losses. Simulation, in this regard, has proved to be an efficient tool to help in decision making purposes. We present the development of a discrete event simulation model for a large-scale open pit mine to run experimental scenarios for near optimal decision making. A mixed integer linear goal programming (MILGP) optimization model is also presented for shovel allocation optimization. The mathematical model will be coupled with the simulation model at the dispatching stage for shovel allocation optimization and thus it will act as the link to optimal short-term schedule to assist in achieving better mine plan compliance across short-term and operational planning time horizons. The main contribution of this work so far is the development of the MILGP model and development, implementation and verification of the simulation model on a large scale oil sands open pit mine case study. The results obtained so far are promising and have been validated against the historical data of the mine, proving the applicability of the model for experimental scenario analysis. INTRODUCTION It is essential for mining operations to adopt efficient operational methods and technologies to cope with market demand and economics. Mining operations, on the other hand, being capital intensive, find it risky to bring in any changes. Simulation is a tool that can help in this regard for better decision making purposes with an improved and quantitative measure of confidence. A true representative model that satisfies the objectives set for the simulation is essential to achieve its applicability for decision making purposes. A similar simulation model is presented in this paper for analyzing the operational KPIs of a large scale oil sand open pit mine with truck-shovel operations. This paper is aimed at providing an overview of the development of the model and the statistical outcomes validating its applicability for experimental scenario runs and thus for better decision making purposes. A mixed integer linear goal programming (MILGP) optimization model is also presented in this paper that acts as an upper stage in a multi-stage dispatching system. The optimization model is proposed to be linked with the dispatching logic in the simulation for optimal shovel assignments and target productions by reading available mining faces directly from the short-term extraction schedule. This model is aimed at bridging the gap between operations and short term extraction schedule; and may help achieve better compliance between the two."