This paper explores the application of sensor data analysis to detect anomalies in temperature measurements of bearings within an industrial centrifugal device used by a mining company. The primary focus of the study was to analyse abnormalities in the temperature data that could indicate potential equipment issues, employing the One-Class Support Vector Machine (SVM) technique for this purpose. Anomaly detection plays a critical role in understanding the behaviour of operational data. It serves as a foundational element in the development of predictive maintenance systems, particularly those aimed at fault classification. In this study, temperature sensors attached to the bearings continuously monitored the device’s operational state, generating a substantial dataset over time. The One-Class SVM technique was applied to this data to identify any deviations from the normal operational patterns, which could signal the onset of mechanical issues or failures. By focusing on detecting these anomalies early, the study aimed to contribute to the creation of a more proactive approach to equipment maintenance, moving from reactive repairs to preventative strategies. The results of the analysis demonstrate that the One-Class SVM is highly effective at pinpointing anomalies, offering valuable insights that can be utilized for the ongoing monitoring and management of equipment health. These insights are instrumental in the proactive surveillance of equipment conditions, ultimately aiding in the development of intelligent maintenance strategies that can reduce downtime, enhance operational efficiency, and extend the lifespan of critical industrial machinery.