[SEO Subhead]
This Guidance illustrates a comprehensive, three-modular approach to connecting and analyzing your industrial equipment, addressing the key stages of equipment connectivity, analytics, and data visualization. It includes an edge location component that enables on-site data ingestion, capturing critical parameters like temperature, vibration, and pressure. The next module shows how to ingest and process real-time data from the edge. It prepares and streams the data to the centralized data lake, where artificial intelligence and machine learning (AI/ML) services use the data to detect abnormal equipment behavior, predict potential failures, and avoid unplanned downtime. Finally, the data visualization module shows how data is ingested from the data lake to provide dashboards and 3D visualizations and enable direct data exploration and analysis.
Please note: [Disclaimer]
Architecture Diagram
Overview
The architecture diagrams for this Guidance are comprised of a suite of three modules: A) Data ingestion, B) Data streaming, processing & machine learning, and C) Data visualization and notifications. This diagram provides a conceptual overview of each module and its interdependencies.
Overview
This architecture diagram consists of three integrated modules that address key stages of equipment connectivity, analytics, and data visualization.
Step A
The customer facility represents the operational site where data is collected from various sources, such as Programmable Logic Controllers (PLCs), SCADA, and Internet of Things (IoT) sensors.
This location also provides connectors for various industrial protocols and edge computing capabilities, including machine learning models for predictive maintenance and asset optimization.
Step B
The data stream module ingests real-time data from industrial equipment and sensors. The data lake establishes a centralized, secure time-series data repository for storing and curating all sensors, machines, production lines, and other industrial equipment.
By consolidating diverse data sources into a single repository, organizations can gain deeper insights into their operations, improve decision-making processes, optimize efficiency, and enable advanced analytics, machine learning, and predictive maintenance.
The artificial intelligence and machine learning (AI/ML) module utilizes advanced algorithms and models to predict equipment failures, optimize production schedules, and improve overall efficiency.
Through anomaly detection, predictive maintenance, and process optimization, machine learning algorithms can identify patterns and trends that may not be apparent through traditional analysis methods.
Step C
The data visualization module presents real-time visibility into key performance indicators, equipment status, and production trends. This enables informed decision-making, proactive maintenance strategies, and optimization of resource utilization.
End users interact with this solution through a visualization dashboard. They also receive notifications and alerts across various devices.
Overview
This architecture diagram consists of three integrated modules that address key stages of equipment connectivity, analytics, and data visualization.
Step A
The customer facility represents the operational site where data is collected from various sources, such as Programmable Logic Controllers (PLCs), SCADA, and Internet of Things (IoT) sensors.
This location also provides connectors for various industrial protocols and edge computing capabilities, including machine learning models for predictive maintenance and asset optimization.
Step B
The data stream module ingests real-time data from industrial equipment and sensors. The data lake establishes a centralized, secure time-series data repository for storing and curating all sensors, machines, production lines, and other industrial equipment.
By consolidating diverse data sources into a single repository, organizations can gain deeper insights into their operations, improve decision-making processes, optimize efficiency, and enable advanced analytics, machine learning, and predictive maintenance.
The artificial intelligence and machine learning (AI/ML) module utilizes advanced algorithms and models to predict equipment failures, optimize production schedules, and improve overall efficiency.
Through anomaly detection, predictive maintenance, and process optimization, machine learning algorithms can identify patterns and trends that may not be apparent through traditional analysis methods.
Step C
The data visualization module presents real-time visibility into key performance indicators, equipment status, and production trends. This enables informed decision-making, proactive maintenance strategies, and optimization of resource utilization.
End users interact with this solution through a visualization dashboard. They also receive notifications and alerts across various devices.
-
Module A: Data ingestion
This architecture diagram displays an edge location component that enables on-site data ingestion from IoT sensors, PLCs, SCADA, and historians.
Click to enlarge
Data Ingestion
The edge location component enables on-site data ingestion from PLCs, historians, industrial equipment, and IoT sensors installed on equipment. This allows the capture of data on critical parameters such as temperature, vibration, and pressure.Step 1
At the edge location, various telemetry data are produced from the PLCs, SCADA, historians, IoT sensors, and industrial equipment.Step 2
Partner connector software solutions act as translation layers, converting data packets from the legacy protocol into a format compatible with a modern standard, such as MQTT or an OPC Unified Architecture (OPC UA).Step 3
AWS IoT SiteWise Edge enables the collection, organization, processing, and monitoring of equipment data on-premises. Local applications that use data from AWS IoT SiteWise Edge will continue to function even during intermittent cloud connectivity.Step 4
The AWS IoT Greengrass machine learning component facilitates inferences locally on devices using models that are created, trained, and optimized in the cloud. This enables the prediction of equipment failure and the avoidance of breakdowns.Step 5
The MQTT broker coordinates the messages between clients, receiving and filtering the messages.Step 6
Custom applications can be created by using IoT Greengrass V2 components to create modular application software.Data Ingestion
The edge location component enables on-site data ingestion from PLCs, historians, industrial equipment, and IoT sensors installed on equipment. This allows the capture of data on critical parameters such as temperature, vibration, and pressure.Step 1
At the edge location, various telemetry data are produced from the PLCs, SCADA, historians, IoT sensors, and industrial equipment.Step 2
Partner connector software solutions act as translation layers, converting data packets from the legacy protocol into a format compatible with a modern standard, such as MQTT or an OPC Unified Architecture (OPC UA).Step 3
AWS IoT SiteWise Edge enables the collection, organization, processing, and monitoring of equipment data on-premises. Local applications that use data from AWS IoT SiteWise Edge will continue to function even during intermittent cloud connectivity.Step 4
The AWS IoT Greengrass machine learning component facilitates inferences locally on devices using models that are created, trained, and optimized in the cloud. This enables the prediction of equipment failure and the avoidance of breakdowns.Step 5
The MQTT broker coordinates the messages between clients, receiving and filtering the messages.Step 6
Custom applications can be created by using IoT Greengrass V2 components to create modular application software. -
Module B: Data streaming, processing, and machine learning
This architecture diagram shows how data from the edge location is processed and ingested into a data lake, along with AI/ML services.
Click to enlarge
Data streaming, processing, and machine learning
This module ingests and processes real-time data from industrial equipment and IoT devices at the edge, preparing and streaming it to the centralized data lake. The AI/ML services then use this data to detect abnormal equipment behavior, enabling the prediction of potential machine failures and the avoidance of unplanned downtime.
Step 7
Use AWS IoT SiteWise to unlock real-time data streaming from industrial equipment, delivering an organized view of live and historical data.Step 8
An industrial data lake using the contextual data provided by AWS IoT SiteWise. Govern, secure, and share data using AWS Lake Formation for advanced analytics. Catalog the data using AWS Glue and analyze the data using Amazon Athena queries.Use the built-in integration of AWS IoT SiteWise and Amazon Lookout for Equipment for anomaly detection.
Step 9
Amazon Monitron is used to detect abnormal equipment behavior so that potential machine failures are detected before failures occur and unplanned downtime is avoided.Custom ML models can be developed with Amazon SageMaker. Amazon Bedrock is used to build and scale generative AI applications with foundation models.
Data streaming, processing, and machine learning
This module ingests and processes real-time data from industrial equipment and IoT devices at the edge, preparing and streaming it to the centralized data lake. The AI/ML services then use this data to detect abnormal equipment behavior, enabling the prediction of potential machine failures and the avoidance of unplanned downtime.
Step 7
Use AWS IoT SiteWise to unlock real-time data streaming from industrial equipment, delivering an organized view of live and historical data.Step 8
An industrial data lake using the contextual data provided by AWS IoT SiteWise. Govern, secure, and share data using AWS Lake Formation for advanced analytics. Catalog the data using AWS Glue and analyze the data using Amazon Athena queries.Use the built-in integration of AWS IoT SiteWise and Amazon Lookout for Equipment for anomaly detection.
Step 9
Amazon Monitron is used to detect abnormal equipment behavior so that potential machine failures are detected before failures occur and unplanned downtime is avoided.Custom ML models can be developed with Amazon SageMaker. Amazon Bedrock is used to build and scale generative AI applications with foundation models.
-
Module C: Data visualization & notifications
This architecture diagram shows how data is ingested and used for dashboards and 3D visualizations.
Click to enlarge
Data visualization & notifications
The data visualization module ingests data from AWS IoT SiteWise and the data lake to provide dashboards and 3D visualizations, as well as enable direct data exploration and analysis from the curated datasets.Step 10
Use AWS IoT TwinMaker to create digital twins, which are virtual representations of the physical operational environment. This enables plant operators to quickly identify and address equipment and process anomalies on the plant floor to improve worker productivity and efficiency.Step 11
Use Amazon Managed Grafana, AWS IoT SiteWise Monitor, AWS IoT App Kit, or an AWS Partner to create a dashboard for the visualization of the digital twin and remotely monitor your equipment in near real-time. The fully-managed service, Managed Grafana, provides rich, contextual dashboards for this purpose.Step 12
Use AWS IoT Events and Amazon Simple Notification Service (Amazon SNS) to monitor the health of industrial equipment for failures or changes in operation and notify the right personnel to take action.Data visualization & notifications
The data visualization module ingests data from AWS IoT SiteWise and the data lake to provide dashboards and 3D visualizations, as well as enable direct data exploration and analysis from the curated datasets.Step 10
Use AWS IoT TwinMaker to create digital twins, which are virtual representations of the physical operational environment. This enables plant operators to quickly identify and address equipment and process anomalies on the plant floor to improve worker productivity and efficiency.Step 11
Use Amazon Managed Grafana, AWS IoT SiteWise Monitor, AWS IoT App Kit, or an AWS Partner to create a dashboard for the visualization of the digital twin and remotely monitor your equipment in near real-time. The fully-managed service, Managed Grafana, provides rich, contextual dashboards for this purpose.Step 12
Use AWS IoT Events and Amazon Simple Notification Service (Amazon SNS) to monitor the health of industrial equipment for failures or changes in operation and notify the right personnel to take action.
Get Started
Well-Architected Pillars
The AWS Well-Architected Framework helps you understand the pros and cons of the decisions you make when building systems in the cloud. The six pillars of the Framework allow you to learn architectural best practices for designing and operating reliable, secure, efficient, cost-effective, and sustainable systems. Using the AWS Well-Architected Tool, available at no charge in the AWS Management Console, you can review your workloads against these best practices by answering a set of questions for each pillar.
The architecture diagram above is an example of a Solution created with Well-Architected best practices in mind. To be fully Well-Architected, you should follow as many Well-Architected best practices as possible.
-
Operational ExcellenceRead the Operational Excellence whitepaper
The services used in this Guidance aid in data ingestion, local processing, event monitoring, message delivery, and machine learning for manufacturing optimization. For example, AWS IoT SiteWise and IoT Greengrass handle data collection, processing, and edge computing, while AWS IoT Events identifies significant sensor data events. Amazon SNS manages messaging, and AI/ML is used to extract insights from operational data for process improvements, predictive maintenance, quality control, and automated inspections.
-
SecurityRead the Security whitepaper
This Guidance uses AWS Identity and Access Management (IAM) and AWS IoT Core to help protect your data, systems, and assets while improving your security posture. For instance, IAM policies are scoped to allow only the minimum required permissions, while IoT Greengrass policies control which devices can send data and interact with the cloud, limiting unauthorized access to prevent accrued charges, tampering, or malicious operations. Additionally, encryption at rest is enabled for all cloud data destinations, like Amazon Simple Storage Service (Amazon S3) and AWS IoT SiteWise. By implementing granular access controls, device policies, and data encryption, these services enhance security by mitigating the risks of unauthorized access, data breaches, and potential threats to industrial assets and operations.
-
ReliabilityRead the Reliability whitepaper
With AWS IoT SiteWise and AWS IoT TwinMaker, throttling limits are in place for data ingress and egress to help assure continued operation. In addition, the Managed Grafana console provides access to a workspace for visualizing and analyzing metrics, logs, and traces without the need to manage infrastructure. It automatically provisions, configures, manages operations, and scales to meet dynamic usage demands—critical for handling peak industrial operations.
For optimal performance, it is recommended that you configure a multi-Availability Zone (AZ) deployment strategy.
By implementing a multi-AZ architecture, throttling safeguards, and managed services that auto-scale, this Guidance helps ensure reliable data ingestion, analysis, and visualization even during high-traffic periods, minimizing downtime and disruptions to industrial processes.
-
Performance EfficiencyRead the Performance Efficiency whitepaper
AWS IoT TwinMaker collects outgoing messages to AWS IoT SiteWise to prevent throttling as tag volumes increase. For analytics, AWS IoT SiteWise and Amazon S3 automatically scale to accommodate data storage needs. For edge applications, AWS IoT SiteWise Edge offers features like edge dashboards and processing optimized for near real-time performance.
In addition, this Guidance uses the auto-scaling capabilities of AWS IoT SiteWise and Amazon S3. It also optimizes the localized processing capability for AWS IoT SiteWise Edge to deliver high performance across data ingestion, storage, analytics, and edge computing. This scalable and optimized approach prevents throttling, accommodates fluctuating data volumes, and provides low-latency processing for time-sensitive industrial operations.
-
Cost OptimizationRead the Cost Optimization whitepaper
For equipment connectivity, the main cost savings come from reduced onsite operational effort, such as regulator costs and human resources. Therefore, you should choose services that bring automation, minimum administration, and scalability. The engineering hours saved should be the primary metric when assessing AWS technologies, as service rates are a smaller part of the financial assessment. Additionally, the total cost of ownership (TCO) and overhead savings are important, as AWS customers often spend significant resources managing on-premises IT infrastructure.
AWS IoT SiteWise and AWS IoT TwinMaker are cost-optimized, managed services that provide digital twin capabilities at the lowest possible price point. Their pay-as-you-go pricing model charges only for the data ingested, stored, and queried. AWS IoT SiteWise also offers optimized storage settings to move data from hot to cold tiers in Amazon S3.
-
SustainabilityRead the Sustainability whitepaper
To address sustainability commitments across industries like manufacturing, construction, energy, utilities, and oil and gas, this Guidance uses AWS IoT SiteWise and Amazon Bedrock. These services enhance sustainability through an elastic IT infrastructure that scales based on usage, minimizing excess compute resources and associated emissions that can be tracked through the Customer Carbon Footprint Tool. Additionally, these services enable engineering agility with digital twins, event-based automation, and AI/ML insights to optimize on-site operations, increase efficiency, and minimize emissions.
AWS IoT SiteWise further supports sustainability with its cold storage tier using the open-source, columnar Apache Parquet format on Amazon S3 for efficient data storage and retrieval.
Lastly, Amazon Bedrock provides data visualization and natural language processing capabilities, where you have access to foundation models to build generative AI applications. These capabilities help you identify unknown risk areas from historical data and validate the effectiveness of interventions, further aiding emission reduction efforts.
Related Content
[Title]
Disclaimer
The sample code; software libraries; command line tools; proofs of concept; templates; or other related technology (including any of the foregoing that are provided by our personnel) is provided to you as AWS Content under the AWS Customer Agreement, or the relevant written agreement between you and AWS (whichever applies). You should not use this AWS Content in your production accounts, or on production or other critical data. You are responsible for testing, securing, and optimizing the AWS Content, such as sample code, as appropriate for production grade use based on your specific quality control practices and standards. Deploying AWS Content may incur AWS charges for creating or using AWS chargeable resources, such as running Amazon EC2 instances or using Amazon S3 storage.
References to third-party services or organizations in this Guidance do not imply an endorsement, sponsorship, or affiliation between Amazon or AWS and the third party. Guidance from AWS is a technical starting point, and you can customize your integration with third-party services when you deploy the architecture.