Savigent Streamline™ accelerates and automates the fault detection process by combining stream processing and complex event processing (CEP). Streamline filters incoming real-time data and events, automatically computing and detecting anomalies.
Detecting Complex Events
Complex Event Processing (CEP) and Event Stream Processing (ESP) are event processing technologies that combine data from multiple sources to infer events or patterns that suggest more complicated circumstances. Using Streamline, meaningful events (such as opportunities or threats) can be quickly identified and responded to. This multi-variable analysis gives manufacturers the capability to infer data that is not directly collected and therefore detect faults or anomalies more efficiently.
Automating Alert Responses
Streamline’s fault detection, when paired with Savigent Workflow™, goes beyond sending an alert to the manufacturing floor (where there may or may not be an operator) by acting immediately (shutting down a machine or making adjustments) when a complex event is detected. This flexibility increases uptime, decreases waste and allows engineers and developers to focus on performing at their highest value.
Multiple Machine Deployment and Reuse
Streamline allows new algorithms and rule sets to be implemented across all the applications executing on the platform. Rule sets can be configured to quickly modify algorithms based upon specific incoming data, resulting in greater efficiency by matching changing targets. While streaming data analysis has become crucial, it is often a manual investigative effort and complicated by multiple data origins. Often engineers end up using multiple applications that replicate data, running iterative reports and wasting crucial technical and human resources. Streamline intelligently aggregates and analyzes multiple data and event streams, eliminating reprocessing of data, and simplifying the fault detection process.
Streamline uses time-encoded streams of sensors, contexts and events. By time stamping each stream separately, Streamline can handle high frequency data, communications delays and late and out-of-order data. This ensures that while data is collected in real-time, analysis and calculations remain highly accurate and can be used to predict future events. Testing of new algorithms can be completed using historical data, captured by Savigent Historian™.