JultikaUniversity of Oulu repository

Abstract

The Kittilä Mine is an underground mine in northern Finland where sulfur content of the ore often restricts gold production. In order to optimize gold production, sulfur content of the ore fed to the mill must remain below a certain limit as to not overload the autoclave process.

An analytical look of the mining process was taken to ensure a solid understanding of the planning process for the mine’s particular mining method. Various sulfur prediction methods were analyzed and using statistical analysis it was determined that the Primary Block Model grade estimates were the most effective grades to use in the remainder of the research. Deswik, the software in use for the mine’s planning and production, was then examined and reviewed from a user’s standpoint. Several aspects of Deswik were tested in attempts to create improved production plans regarding the sulfur limits or improved NPV. Throughout the trials, plans which included improved short term results delayed important development through the mine and disrupted steady production through the long term schedule. Other issues were found through trials, including the tendency of the optimization algorithms to take advantage of broken links throughout the task-progression network. The various methods and the corresponding results were compared and the advantages and disadvantages of the Deswik system were assessed.