Real-time detection uses streaming data by comparing previously seen data points to the last data point to determine if your latest one is an anomaly. This operation generates a model using the data points you send, and determines if the target (current) point is an anomaly. By calling the service with each new data point you generate, you can monitor your data as it's created.
Real-time detection example
Consider a scenario in the carbonated beverage industry where real-time anomaly detection may be useful. The carbon dioxide added to soft drinks during the bottling or canning process needs to stay in a specific temperature range.
Bottling systems use a device known as a carbo-cooler to achieve the refrigeration of the product for this process. If the temperature goes too low, the product will freeze in the carbo-cooler. If the temperature is too warm, the carbon dioxide will not adhere properly. Either situation results in a product batch that cannot be sold to customers.
This carbonated beverage scenario is an example of where you could use streaming detection for real-time decision making. It could be tied into an application that controls the bottling line equipment. You may use it to feed displays that depict the system temperatures for the quality control station. A service technician may also use it to identify equipment failure potential and servicing needs.
You can use the Anomaly Detector service to create a monitoring application configured with the above criteria to perform real-time temperature monitoring. You can perform anomaly detection using both streaming and batch detection techniques. Streaming detection is most useful for monitoring critical storage requirements that must be acted on immediately. Sensors will monitor the temperature inside the compartment and send these readings to your application or an event hub on Azure. Anomaly Detector will evaluate the streaming data points and determine if a point is an anomaly.
Top comments (0)