Application of the Metaheuristic Approaches in Open Pit Mine Planning

摘要

Production planning of the open pit mines has been one of the challenging mathematical problems for mine engineers in the last decades. It is usually a large optimization model which is impossible for solution in a reasonable period of time even in its simplest formulation. The problem deteriorates by the recent tendencies for incorporation of the existing complexities such as nonlinearities in parameters, uncertainties of the ore grades and doubts about economic and marketing factors. Metaheuristic approaches such as genetic algorithm, simulated annealing, ant colony optimization, particle swarm optimization and tabu search are a set of methodologies that have shown their capabilities in solution of the large and complex problems through severe investigations and numerous researches. The paper summarizes the results of the research studies that have been accomplished by the authors on application of the metaheuristics including genetic algorithm, ant colony optimization and particle swarm optimization in production planning of the open pit mines. It has been proved that metaheuristics could be considered as a new set of open pit optimization methods and their required memory and amount of computations are not as much of mathematical programming. Although the global optimum is not guaranteed in metaheuristics, however, NPV of the mine plans produced by some commercial packages have been improved in most cases from 2 to 8 percent. Procedure of the grade uncertainty incorporation into long-term planning has also been explained and the results have been discussed.