Using Linear Programming to Minimize Freshwater Use in the Gold Processing Industry

Abstract

Gold mines deliver gold ore and waste rock to processing facilities, after which the ore is separated from the rock through an 11-stage process. The stages of ore processing require large quantities of fresh-water which are drawn from nearby lakes or rivers. Given the greater importance placed on environmental sustainability, gold producers have become increasingly interested in reducing the amount of freshwater required to process their ore. One strategy by which this can be achieved is by replacing fresh-water with recycled water, wherever feasible, within the 11 stages of processing. The objective of this research is to develop and apply an optimization model by which the gold processing industry can reduce its use of fresh-water by identifying, within the processing stages, where, and up to how much recycled water can replace fresh-water. To achieve this objective, a linear programming model of this optimal water allocation problem was developed to minimize the use of fresh-water in ore processing, subject to maintaining the feasibility of the processing stages by satisfying constraints on pollutant concentrations. The model was applied to a gold processing facility owned by Goldcorp Ltd. in Red Lake, Ontario, Canada. The results show that the optimal solution generated by the model required 51 metric tonnes/hr of fresh-water versus the current use of 68.6 metric tonnes/hr – a reduction in freshwater use of 25.7%. This research is innovative insofar as an optimization model aimed at minimizing fresh-water usage has not been applied to the gold processing problem by prior researchers.