The purpose of this paper is twofold: an approach is proposed to determine the optimum replacement time for shovel teeth; and a risk-quantification approached is developed to derive a confidence interval for replacement time.
The risk-quantification approach is based on a combination of Monte Carlo simulation and Markov chain. Monte Carlo simulation whereby the wear of shovel teeth is probabilistically monitored over time is used.
Results show that a proper replacement strategy has potential to increase operation efficiency and the uncertainties associated with this strategy can be managed.
The failure time distribution of a tooth is assumed to remain “identically distributed and independent.” Planned tooth replacements are always done when the shovel is not in operation (e.g. between a shift change).
The proposed approach can be effectively used to determine a replacement strategy, along with the level of confidence level, for preventive maintenance planning.
The originality of the paper rests on developing a novel approach to monitor wear on mining shovels probabilistically. Uncertainty associated with production targets is quantified.
