Industrial IoT (IIoT) bridges the gap between industrial equipment and automation networks (usually called OT, Operations Technology) and Information Technology (IT). In IT, use of new technologies such as machine learning, cloud, mobile, and edge computing are becoming commonplace. IIoT brings machines, cloud computing, analytics, and people together to improve performance, productivity, and efficiency of industrial processes enabling customers to turn to IIoT applications for predictive quality and maintenance and to remotely monitor their operations from anywhere.
But realizing the value of IIoT is challenging and usually there are three elements that, from our experience, are holding manufacturers back:
- Data is collected too infrequently
- Data is difficult to access
- Unable to overcome data silos and link data together
This blog will explore how industrial companies can use Predictive Quality to determine actions such as adjusting machine settings, using different sources of raw materials, or doing additional worker training that will improve the quality of the factory output.
By leveraging AWS IoT services, industrial companies across different industries such as mining, energy and utilities, manufacturing, precision agriculture, and oil and gas can reason on top of operational data and improve performance, productivity, and efficiency.
Current situation and challenges of the Industry
Whether you are in manufacturing, utilities, mining, oil and gas, or any other industrial market segment, you have legacy equipment that has been working reasonably well for the last 10, 20 or even 30 years. Many industrial companies have made significant investments in operational technologies such as Industrial PCs (IPCs), Programmable Logic Controllers (PLCs) or large Distributed Control Systems (DCS) with real-time distributed control networks (fieldbuses) connecting them, and Supervisory Control and Data Acquisition (SCADA) systems. Operational technologies were designed and deployed to last a couple of decades, are deeply entrenched, and are very difficult to replace.
The following picture shows the ISA-95 Industrial Edge Architecture and how the above-mentioned elements are related to each other.
Figure 1 – The automation pyramid according to the ISA 95 model (source: researchgate.net)
If we want to benefit from new technologies like IoT, machine learning, and computer vision, we have to adapt existing equipment and systems, which were not designed for IoT applications.
First challenge for any IIoT application is to connect legacy equipment, so we are able to collect data from different devices (sensors, actuators, electrical motors) from different manufacturers. In many situations we have to adapt to different industry protocols or even retrofitting, by adding new technology to older systems so they are able to measure, allow remote control, and connect.
The second and most important challenge comes together with the connectivity, is security. We have to keep the device and its data secure. The failure of an equipment or system in a production environment can result in costly downtime and impact to the business. We have to ensure that industrial connected devices are able to operate at top performance without cloud connectivity. Data collection processes must not interfere with the operation of the device and ensure that any remote control or update operation is done in a secure way from only allowed operators.
Once we have the data secured comes the third challenge, to get insights. Data can be locked in different “floors” of the factory (different levels of the ISA-95 architecture). In order to get insights from all the raw data, it is key to link the data together, no matter if the data is coming from different devices, manufacturers, historians, fieldbuses, systems, or databases.
How it works
AWS IoT helps industrial companies overcome challenges to attain business goals.
First, AWS IoT lets you easily connect, manage, and update devices of any type from small microcontrollers, to more powerful gateway devices. You can integrate your existing legacy equipment on the manufacturing floor such as Programmable Logic Controllers (PLC) and Supervisory Control and Data Acquisition (SCADA) systems by deploying simple sensors to monitor processes and track key performance indicators without overhauling or replacing existing hardware.
Second, AWS IoT provides built-in device authentication and authorization to keep your IoT data and devices protected. You can also continuously audit security policies associated with your devices, monitor your device fleet for abnormal behavior, and receive alerts if something doesn’t look right. You can even take corrective actions, such as powering off devices or pushing a security fix.
Third, AWS IoT enables connected devices to operate with intermittent Internet connectivity to mitigate risks of unexpected downtime. You can run machine learning models or software code and store data locally until Internet connection is available.
AWS IoT provides “plug and play” capabilities so you can scale your IoT applications to thousands or millions of devices. With AWS IoT, you can organize device inventory, monitor your fleet of devices, and remotely manage devices across many locations including updating device software over-the-air (OTA).
In the next picture you can see how the different AWS IoT Services can work together to make IIoT a reality for your business.
Figure 2 – AWS IoT Industrial Reference Architecture.
Once devices are securely onboarded, AWS IoT offers an easy way to run analytics on IoT data. AWS IoT collects, processes, and analyzes IoT data quickly and easily so you can gain operational insights. AWS IoT integrates with Amazon SageMaker so you can build machine learning models for your Industrial IoT Data. These machine learning models can run in the cloud and can be deployed locally on devices. With Amazon QuickSight, you can visualize and explore data and share insights across teams.
In the following section, you will find a detailed explanation how the different AWS IoT Services provide value to support the most important industrial use cases:
- Asset Condition Monitoring
- Predictive Maintenance
- Predictive Quality
Top Industrial Use cases and architecture walkthrough
Asset Condition Monitoring
Asset condition monitoring captures the state of your machines and equipment so you can understand how the asset is performing in the field or on the factory floor. Typically, data such as temperature, vibration, and error codes indicate if equipment usage is optimal but it’s hard to capture manually since technicians need to physically inspect machines. With AWS IoT, you can capture all IoT data and monitor performance. With increased visibility, you can maximize asset utilization and fully exploit your investment.
We suggest the following reference architecture for doing Asset Condition Monitoring (aka Condition Based Monitoring) in industrial environments.
Figure 3 – AWS IoT Industrial Reference Architecture for Asset Condition Monitoring
Let’s review the role of each AWS IoT Services to provide you a holistic Asset Condition Monitoring use case inside your industry:
- AWS IoT Greengrass brings local compute, messaging, data caching, sync, and ML inference capabilities to edge devices. In this specific use case, AWS IoT Greengrass provides you:
- With AWS IoT Greengrass Connectors and long-running lambda functions to integrate with any existing industrial protocols and devices. Besides that, with AWS IoT Greengrass you have also access to the local resources of the gateway where it is running, enabling AWS IoT Greengrass to receive sensor data and manage device via GPIO, serial ports or any other interface. AWS IoT Greengrass can connect with higher-level systems like SCADA or even MES to enrich the information coming from the industrial devices and also to feed data back from the shop floor to the MES bus. In the following picture you can see how AWS IoT Greengrass can communicate with existing legacy devices.
Figure 4 – AWS IoT Greengrass doing Protocol Translation to connect to existing factory machines.
- You can also operate offline. AWS IoT Greengrass lets connected devices operate even with intermittent connectivity to the cloud. Once the device reconnects, AWS IoT Greengrass synchronizes the data on the device with AWS IoT Core, providing seamless functionality regardless of connectivity.
- Helps you reduce the cost of running IoT applications. You can get rich insights at a lower cost by programming your device to filter data locally (and even doing machine learning inference at the edge) and only transmit the data you need for your applications to the cloud. This reduces the amount of raw data transmitted to the cloud, minimizing cost and increasing the quality of the data you send to the cloud. You could even have the ETL paradigm (Extract-Transform-Load) at the edge, where you extract the data from the factory machines doing protocol conversion, you transform the data into the right format and then load (i.e. send) the data into AWS IoT Core.
Figure 5 – AWS IoT Greengrass doing local processing of IoT data.
- AWS IoT Core is a managed cloud service that lets connected devices easily and securely interact with cloud applications and other devices. AWS IoT Core can support billions of devices and trillions of messages, and can process and route those messages to AWS endpoints and to other devices reliably and securely. In this specific use case,
- AWS IoT Core can filter, transform, and act upon device data on the fly, based on business rules you define. You can use the IoT Rules to be able to detect in real-time malfunctioning in equipment and redirect this information to the right service. In this case, if errors are detected, we send those errors to the AWS SNS messaging service to send either a SMS or an Email to the factory manager to take actions. Besides that, all the information is sent to AWS IoT Analytics for further processing and analyzing of the data.
- AWS IoT Analytics is a fully-managed service that makes it easy to run and operationalize sophisticated analytics on massive volumes of IoT data without having to worry about the cost and complexity typically required to build an IoT analytics platform. In this specific use case,
- AWS IoT Analytics can enrich the IoT data received from the industrial equipment with information located in other sources, can fill the gaps if data is missing, can eliminate false readings and can perform mathematical operations in case sensors are not right calibrated.
- AWS IoT Analytics can prepare the data to be visualized directly with Amazon QuickSight and to be analyzed with machine learning using Amazon SageMaker.
With the basis of this architecture we have the right pieces to address the following use case, predictive maintenance.
Predictive maintenance analytics captures the state of industrial equipment so you can identify potential breakdowns before they impact production. With AWS IoT, you can continuously monitor and infer equipment status, health, and performance to detect issues in real-time. When organizations use predictive maintenance analytics, equipment lasts longer, worker safety increases, and the supply chain is optimized.
For predictive maintenance we suggest the following reference architecture:
Figure 6 – AWS IoT Industrial Reference Architecture for Predictive Maintenance
If you compare this architecture with the previous one for Asset Condition Monitoring you will see a couple of extra functionalities that enable you to anticipate equipment failure:
- Amazon SageMaker is a fully-managed service that covers the entire machine learning workflow to label and prepare your data, choose an algorithm, train the model, tune and optimize it for deployment, make predictions, and act. In this specific architecture,
- Amazon SageMaker can directly apply already existing or any custom-built algorithm on top of the clean data processed by AWS IoT Analytics. You can do statistical classification through a method called logistic regression. You can also use Long-Short-Term Memory (LSTM), which is a powerful neural network technique for predicting the output or state of a process that varies over time. The pre-built notebook templates also support the K-means clustering algorithm for device segmentation, which clusters your devices into cohorts of like devices. These templates are typically used to profile device health and device state such as HVAC units in a factory or wear and tear of blades on a wind turbine.
- AWS IoT Greengrass makes it easy to perform machine learning inference locally on devices, using models that are created, trained, and optimized in the cloud. AWS IoT Greengrass gives you the flexibility to use machine learning models in Amazon SageMaker or to bring your own pre-trained model stored in Amazon S3. In this architecture,
- Once the predictive model is trained in AWS Cloud, the model can be deployed in AWS IoT Greengrass and perform machine learning inference locally. In such way, you can run immediate corrective actions on the edge, locally, if the predicted model anticipates malfunctioning behavior, then your factory will run always on the safe side.
Now, let’s enhance the predictive maintenance reference architecture to reach predictive quality, which is the goal of any smart factory.
Predictive quality analytics extract actionable insights from industrial data sources such as manufacturing equipment, environmental conditions, and human observations. The goal of predictive quality analytics is to determine actions such as adjusting machine settings or using different sources of raw materials that will improve the quality of the factory output. Using AWS IoT, industrial manufacturers can build predictive quality models which help them build better products. Higher quality products increase customer satisfaction and reduce product recalls.
The recommended reference architecture for predictive quality, not only monitors the state of the industrial equipment (asset condition monitoring) and predicts failures (predictive maintenance), but also monitors the quality of the manufactured product during all the steps of the production line, by adding computer vision and machine learning as you can see in the following picture:
Figure 7 – AWS IoT Industrial Reference Architecture for Predictive Quality
In this new architecture we add the following elements:
- Computer Vision to capture via images and/or videos the product in each of the phases. In this architecture,
- Thanks to AWS IoT Greengrass you can connect to any simple camera, perform the required protocol translation, and transform that camera in a smart camera by running machine learning inference at edge.
- Initially, enough images and/or videos have been uploaded to the cloud and stored in S3, to be able to train a vision machine learning model appropriate to your product. This model will do the detection of faulty products automatically.
- Once the machine learning model is trained, we can deploy this model in AWS IoT Greengrass, and run machine learning inference locally, so even if you lose connectivity to internet, you will be still able to do the inference and asses the quality locally.
When industrial companies get started building IoT applications, they are often concerned about onboarding legacy equipment. data and device security, unexpected downtime, and getting valuable insight from collected data.
AWS IoT provides “plug and play” capabilities to connect with existing equipment, no matter what manufacturer, extract data from devices and control them in a secure way and finally to get the needed insights with all this data.
In this blog we have reviewed the recommended AWS IoT services and architectures to do Asset Condition Monitoring, Predictive Maintenance, and Predictive Quality to enable you to easily digitize your industry.