Challenges and Cases of Artificial Intelligence Applied to Assist Predictive Maintenance in the Industry, Respectively in the Mining Sector

<p>Unscheduled maintenance events have a crucial role in the industry's value chain. Predictive maintenance is<br>becoming notorious in the mining sector because of improving downtime planning, operation effectiveness, and<br>asset reliability. Although data analysis is widely used in other sectors, it is still necessary to deepen studies and<br>applications for the mining industry. Therefore, the objective of this study is to identify, in the current state of<br>the art, the existing challenges and solutions for Machine Learning techniques applied as an aid for predictive<br>maintenance and, finally, present recent studies that used Artificial Intelligence in the mining industry to aid<br>maintenance. The challenges reflect the natural world's complexity, such as dealing with limited, imbalanced,<br>and noisy data, real-time extensive data monitoring, identifying relevant predictors, and ensuring data security.<br>Existing solutions comprehend Machine Learning, Artificial Intelligence, and other techniques used in Industry<br>4.0.</p>