Predictive Maintenance in Underground using artificial intelligence (#2114)

Authors

Chambi, Nelson

Sanga, Celso

Sanga, Alejandra

Sanga, Piero

Abstract

The article addresses predictive maintenance (PdM) applied to underground mining equipment using artificial intelligence, a crucial approach for improving efficiency and reducing operating costs. The objective is to optimize the equipment's lifespan through early fault detection, avoiding costly repairs and unplanned downtime. The challenge lies in the extreme conditions and intensive use of the equipment, which makes it difficult to predict failures using traditional methods. The methodology includes continuous monitoring of key parameters (temperature, pressure, oil analysis, thickness measurement) through sensors and real-time data analysis. This data is processed using artificial intelligence and machine learning techniques to identify patterns that precede failures. The results show that PdM can reduce maintenance costs by 8% and increase equipment availability by 10%, leading to greater productivity and safety in underground mining operations