AWS Industrial PdM ML Model and Anomaly Detection
Publication date: March 4, 2024 (Diagram history)
With this architecture, you can create a predictive maintenance (PdM) ML model by using Amazon SageMaker AI with AWS IoT Core and an anomaly detection application by using Amazon Managed Service for Apache Flink. This solution uses AWS IoT Greengrass, Amazon Data Firehose, Amazon Simple Storage Service, and Amazon Simple Notification Service.
Industrial PdM ML anomaly detection architecture diagram
The following steps describe the architecture:
-
Configure AWS IoT Greengrass with connectors to communicate with factory machines.
-
Configure rules in AWS IoT Core to trigger events based on Message Queuing Telemetry Transport (MQTT) topics.
-
Create an Amazon Data Firehose delivery stream to store factory data in an Amazon S3 data lake.
-
Build a PdM ML model with SageMaker AI.
-
Deploy the ML model onto the AWS IoT Greengrass Edge Gateway.
-
Build data queries in Amazon Athena against the AWS Glue Data Catalog.
-
Visualize analysis with AWS Quick Sight on the Athena data source.
-
Create an anomaly detection application in Amazon Managed Service for Apache Flink.
-
Configure AWS Lambda as output to send anomaly notifications to Amazon SNS.
For related reference architectures, see AWS Industrial PdM ML Model with Modbus Communication and AWS Industrial IoT Predictive Maintenance ML Model.
Further reading
For additional information, see the following resources:
Diagram history
To be notified about updates to this reference architecture diagram, subscribe to the RSS feed.
| Change | Description | Date |
|---|---|---|
Initial publication | Reference architecture diagram first published. | March 4, 2024 |
RSS subscription
To subscribe to RSS updates, you must have an RSS plugin enabled for the browser that you are using.
