Amazon Lookout for Equipment uses historical data and maintenance records (optional) from your existing machinery sensors to create a unique machine learning (ML) model for you to detect abnormal equipment behavior.
Sensor and data quality evaluation
Time series data coming from sensors on industrial equipment can be highly erratic and the quality/usability of each sensor is difficult to determine. Amazon Lookout for Equipment will derive key statistics on ingested data from each sensor, grade the overall data quality and give a justification for its grade. This output advises a user on which sensors are preferred inputs.
Automated machine learning
Amazon Lookout for Equipment will automatically leverage data from up to 300 sensors at once, as well as maintenance history, in order to search through up to 28,000 possible algorithm combinations and determine the optimal multi-variate model that best learns the normal behavior of the specified equipment.
For each detected abnormal behavior, Amazon Lookout for Equipment will understand the behavior and indicate to a user which sensors are impacting the issue and what is happening in each of those sensors. Customers can use this information to diagnose the problem and take corrective action.
Easily deploy the developed model on real time data by setting up an inference scheduler. The scheduler will run inferencing on newly generated sensor data at intervals as low as once per minute and as high as once per hour.