Operational Anomaly Detection

AI monitors operational data to surface anomalies, issues, and early warning signals in ongoing operations.

Business impact

• Tasks automated — Ericsson reports 10,000 tasks automated in its operations transformation.
• Automation rules reuse — Ericsson reports 6,000 automation rules with 85% reuse.

Data requirements

• Laboratory operational metrics (tabular) — Recursion analyzes operational metrics from lab operations.
• Instrument monitoring data — Recursion uses data tied to monitoring and diagnosing lab instruments.
• Network data (tabular) — Ericsson processes network data in managed services delivery operations.

AI methods and techniques

• Anomaly detection — Used to surface unusual patterns in laboratory operational metrics and operational data.
• Machine learning — Used in the Splunk Machine Learning Toolkit and in Ericsson’s operations engine to analyze operational data.
• Predictive analytics — Used in Ericsson’s managed services operations to anticipate failures.

Industries

Real-world evidence

2 documented case studies on record.

Companies using this: Recursion, Ericsson.