Project Area Description
In this section we present the projects developed by ARAMIS for the Process industry.
We developed a novel approach for tracking a global risk index of a safety-critical industrial plant. This approach, called Integrated Risk Index by Sapio (IRIS), considers safety as a critical process variable and relies on a combination of techniques such as Bayesian Networks, Multi Attribute Value Theory, Cognitive Reliability and Error Analysis Method to build an on-line index, encoding the whole variety of risk-related factors (including the human factor).
The IRIS approach produces an overal risk index, which simplifies risk communication and aids decision making. The on-going steps for the IRIS enhancement are the development of interfaces with the Enterprise Asset Management (EAM) software, to trigger alarms with work requests and/or work orders, thus enabling a prescriptive maintenance approach.
Our client monitors hundreds of wind turbines of different technologies, and needed an automatic system to detect anomalies. To address this issue, Aramis has developed a diagnostic tool based on Isolation Forests. This is incrementally built from first-attempt, rough solutions based on the limited knowledge, information and data on the monitored system available at the beginning of the development.
Many sources of operational data and health indexes were first aggregated and streamlined to be available in production, an online environment. Finally, an alert system was developed in coordination with the business department.
The developed diagnostic tool helps decision making on wind turbine operation and maintenance, granting a massive gain in power production.