Figure 24 - Integration Model 3.1. Technological Architecture • IOT: Implementation of smart sensors and telemetry devices across critical assets (Shovels, Trucks, Drills). These devices capture real-time health variables (temperature, pressure, vibration) and PLC logic states to detect failures like "Motor Stalled" before they result in significant downtime. • Big Data & Cloud Computing: Utilization of cloud infrastructures (Azure/AWS) for the massive storage of high-frequency and historical data. This allows for cross-referencing information between different fleets and mine sites, eliminating information silos and enabling scalable processing. • Visualization: Development of interactive dashboards (Embebedd BI/Grafana) that translate complex data into visual KPIs, such as the "Quantity vs. Duration" charts previously analyzed, allowing supervisors to prioritize interventions in seconds with time series information. Figure 25 - Visualization App 3.2. Data Analytics and Predictive Models • Base Analytics and Automated Reporting: Migration from manual Excel-based reporting to automated systems that generate daily/monthly failure summaries. This reduces human error and frees up time for critical analysis. • Machine Learning: The Stress Cycle Modeling The algorithm processes real-time variables such as hoist motor torque, bail pull, and payload per cycle to calculate the cumulative fatigue on the hoist cables. By analyzing the "current signatures" of the hoist motors (linked to the HP1 and H2 motors mentioned in the reports), the model identifies micro-stretches or irregular vibrations that precede a strand failure.(Göger et al., 2026)
RkJQdWJsaXNoZXIy MTM0Mzk2