Summary
The project develops an Intelligent Failure Prediction System that transforms the traditional maintenance strategy (corrective and scheduled preventive) towards a predictive, conditionbased approach. In response to increasing demands for operational reliability, the solution integrates multiple information sources to generate a dynamic risk ranking, enabling the prioritization of preventive interventions and optimizing system reliability.
Read more Read lessStrategic Context: The organization manages a tripartite maintenance strategy: corrective, preventive, and condition-based. The demand for higher reliability has motivated the development of tools that enhance predictive maintenance, which has traditionally faced resistance to change and established operational practices.
Main Innovation: The system uses an intelligent association algorithm that correlates operational events (forced events, emergency maintenance, unplanned maintenance) with specific equipment based on evidence from maintenance notifications, ratings, and unavailability, overcoming the limitations of traditional methods that rely solely on direct equipment-event associations. This tool is part of a macro MLOps (Machine Learning
Operations) ecosystem that seeks to answer 'the why of things' from a comprehensive data science, data engineering, and artificial intelligence perspective.
Additional informations
| Publication type | Session Materials |
|---|---|
| Reference | A3_12191_2026 |
| Publication year | |
| Publisher | CIGRE |
| Country | Colombia |
| Study committees | |
| File size | 942 KB |
| Price for non member | 30 € |
| Price for member | 30 € |
Authors
CANTOR Elkin - Intercolombia; PATIÑO Alexander - intercolombia; ARIAS Andrés - Intercolombia
