GreeceEnergy and utilities2021 – ongoingMotor Oil Group
Machine-learning predictive maintenance for refinery compressors
Motor Oil's compressor-monitoring pilot explained historical abnormal events with up to 77% accuracy, SAP reported
Problem
Scheduled maintenance reviews could leave equipment problems undetected between inspections. Motor Oil wanted earlier warning of abnormal compressor behavior.
Approach
Motor Oil worked with SAP and Accenture on a predictive-maintenance pilot. Models analyzed data from 86 sensors on 5 critical compressors and ran hourly on SAP Business Technology Platform. Dashboards and email notifications delivered the results.
Results
Takeaway
The accuracy result concerned explanations of historical abnormal events. SAP did not quantify achieved maintenance savings or downtime reductions.
Categories
- Industry
- Energy and utilities
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Partner co-development
- Evidence
- Vendor case study
Sources
Source summaries
SAP described a completed compressor-monitoring pilot built with Motor Oil and Accenture. It reported up to 77% accuracy explaining historical abnormal events. Achieved maintenance savings and downtime reductions were not quantified.
Full details are in the original sources.
