AWS Industrial IoT Predictive Maintenance ML Model
Publication date: July 8, 2020 (Diagram history)
With this architecture, you can create a Predictive Maintenance (PdM) machine learning (ML) model by using AWS IoT SiteWise and AWS IoT Analytics. AWS IoT SiteWise collects, organizes, and stores data from factory equipment. This makes clean, contextual, and structured data sets available for data scientists to train ML models.
Predictive maintenance ML model architecture diagram
The following steps describe the architecture:
-
Configure the AWS IoT SiteWise Connector on AWS IoT Greengrass to connect and collect data from factory machines by using OPC Unified Architecture (OPC-UA).
-
Use AWS IoT SiteWise to model assets that represent on-premises devices, equipment, and processes.
-
Create a custom web portal with AWS IoT SiteWise Monitor to visualize factory data in near real time. IAM Identity Center provides user authentication for the portal.
-
Use the AWS IoT Core Rules Engine to route data to AWS IoT Analytics.
-
For other industrial data, use AWS IoT Greengrass stream manager to publish data to AWS IoT Core.
-
Build a Docker image and add it to Amazon Elastic Container Registry (Amazon ECR).
-
In AWS IoT Analytics, create a Container Data set from the AWS IoT SiteWise Data store. Link it to the Docker container in Amazon ECR.
-
Create a Jupyter Notebook for the data set to build a PdM ML model. Use Amazon SageMaker AI for model training and deployment.
-
Visualize analysis with Amazon Quick Sight on the AWS IoT Analytics data source.
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. | July 8, 2020 |
RSS subscription
To subscribe to RSS updates, you must have an RSS plugin enabled for the browser that you are using.
