Abstract
Capital intensive industrial assets require highly specialized maintenance activities. Traditional preventive time-based approach, based on OEM maintenance policies, has been gradually evolving towards more sophisticated condition-based maintenance techniques. Further ISO 55000 states that assets exist to provide value to the organization and its stakeholders (BS ISO 55002, 2014). To develop a successful and modern maintenance program, it suggests having a value-based approach when dealing with maintenance decisions, both financial and non-financial constrains needs to be evaluated when decision taken regarding maintenance actions. Higher values can be reaped from an asset when the maintenance intervals are optimized. By optimization it is envisaged that the right number and type of maintenance tasks, at the right intervals, in the right way is performed on the asset to maximize the risk reduction within available budgetary constraints.
The paper presents an overview of an analytics framework for predictive maintenance service boosted by Machine Learning and asset knowledge, applied to turbomachinery assets. Optimization of the maintenance scenario is performed through a risk model that assesses online health status and probability of failure, by detecting functional anomalies or aging phenomena and evaluating their impact on asset serviceability.
