Anomaly detection allows reliability engineers to identify and monitor unusual patterns or behavior in technical objects. It automatically recognizes inconsistent patterns in sensor data streams and in turn significantly reduces human efforts of visual data inspection. The AI-based algorithm is initially fitted to a normal condition of the technical object using indicator data. An assessment is run daily which compares the data set of the last days with the normal condition and raises an alert in the Alerts application when the data differ sufficiently from the normal condition.

Benefits

• Proactively schedule maintenance and boost productivity
• Make advanced data analysis accessible to everyday maintenance staff
• Significantly reduce manual work of reliability engineers