Predictive maintenance in mining industry: grinding mill case study

Abstract

The maintenance of industrial mining machines is a challenging process, especially in the case of critical assets, as it impact the hole process: for instance the grinding mills. However, many approaches have been developed by the growth of the mining industry in order to reduce the cost and optimize the time. Traditionally, machines operate until a failure occurs and then process to the maintenance (curative maintenance). However the loss of equipment and the time squandered during the maintenance procedure, rather than the stoppage of production, make this case inefficient. The second case is preventive maintenance, where the condition of the equipment is not taken into account, thus even healthy components are also susceptible to be replaced, which is a waste of time and budget. This is where predictive maintenance appears. This work aims to present a state of the art of recent works in predictive maintenance using machine learning algorithms including deep learning algorithms to end up with a synthesis of all cases that use predictive maintenance, and propose a Data Driven Model for mining industry assets for a better predictive maintenance.