Applications of Edge and Cloud: The Future of Industrial IoT

We hear a lot today about concepts like the Industrial Internet of Things, or IIoT, edge computing, and cloud computing.

If you are like many automation engineers, these concepts may seem like complex and costly approaches to automation best left for vendors, large systems integrators, and those with advanced degrees. Nothing could be farther from the truth.

If you're new to IoT or want to learn the basics, check out our course, Internet of Things (IoT): Learn the Basics.

In this article, I will demonstrate how edge and cloud computing can be applied to your manufacturing process or plant using some practical examples.

Let’s start with a couple of definitions. IIoT can be defined as a network of smart devices that communicate with each other and can optimize industrial processes without human intervention.

IIoT facilitates the flow of data from the plant floor to applications that can reside in a computer or device inside the plant or offsite in a data center. The concept of IIoT is simple: send data to where it is needed to facilitate real-time manufacturing decisions.

The data produced on the plant floor can be analyzed locally, with devices called edge devices, or in larger data centers, typically in what is called the cloud.

If real-time logic processing or data analysis is required, edge devices can provide a fast, secure path for collecting data from the plant floor. Edge computing devices provide logic processing, alarm handling, and other services that make critical decisions to help improve operations.

Cloud computing is a similar approach, except data processing and analysis are handled by off-site data centers. Large cloud environments can span multiple data center locations, giving the user almost unlimited access to computing power.

This allows for larger applications to be executed, but the time between sending data and receiving output from the cloud application is typically greater than what it takes to receive a response from an edge computing device.

Cloud environments have more computing resources and storage capacity than edge devices, but a user typically pays a recurring fee for using these resources.

What is edge computing?

As mentioned before, edge computing allows for plant data to be processed closer to the source.

Edge devices are computing boxes that are typically connected to the same in-plant network as the source of this data. Therefore, latency, or the time it takes to receive, process, and return results is very small.

Because data remains inside the plant networks, data security is more easily maintained.

Edge computing devices are also connected to the internet and to cloud computing systems to pass information required for additional analysis and processing.

Because not all data collected from the plant floor need to be passed to the cloud, aggregation of data at the edge device reduces the bandwidth necessary to communicate this data to the cloud. This communication to and from the cloud occurs over the internet.

Edge devices provide enhanced reliability since they can continue to operate if communication with the internet is lost. As the plant’s data and processing needs increase, additional edge devices can be added easily to provide the necessary increase in capacity.

We talked more about how smart sensors and actuators can communicate with edge devices in the course, IIoT Fundamentals: Smart Sensors & Actuators in Automation. Complete that course and earn a certificate from Emerson.

What is industrial cloud computing?

Cloud computing provides an enhancement to edge computing, not a replacement or substitute for edge computing. You could, but likely would not, control devices on the plant floor with applications based in a cloud environment.

Cloud computing provides larger data storage capabilities and increased processing power beyond what in-plant systems can provide.

Cloud environments also provide advanced computing capabilities as is needed for machine learning and artificial intelligence. Also, cloud computing allows a company to have worldwide access to this data storage and computing infrastructure.

Integration of edge and cloud for automation

As mentioned, edge computing and cloud computing work together.

Data that needs to be processed locally, such as alarming and control decisions, are made in edge computing devices. Data that is required for complex calculations, such as predictive maintenance or AI applications, is passed to the cloud for processing.

IIoT is dependent on both approaches to provide fast, reliable data analysis on the plant floor while providing resources in the cloud for complex analytics.

By working together, edge computing and cloud computing provide a complete range of analytical applications, data storage, and data security capabilities for any sized company.

A practical example

Do you now have a better understanding of edge and cloud computing and how they function together in an IIoT environment? Good. Let’s look at a practical example that could be employed in your manufacturing plant.

Here is a typical manufacturing plant that bottles a juice product. This is similar to a project I completed not too long ago. Data is critical to this process.

The plant records details of how the product was produced, detailed regulatory compliance information, process control information such as control loop performance, and records of manual operator additions and interactions with the process.

In the end, the plant needs to produce several reports and analyze several trends and parameters. Not all of these analyses can be made locally, and some of these analyses need to occur quickly and in real-time to ensure the purity and consistency of the product.

The system was ultimately constructed with the following items processed at the edge:

• Batch Reporting
• Alarming
• Data Historian
• Control performance

These items were processed in the cloud:

• Batch-to-batch comparisons for consistency
• Detailed analysis of batch times, throughput, and efficiency
• Converting raw material and labor usage to a batch cost

For the batch report created at the edge, data from the plant floor and the cloud computations are used in the final report.

Notice a couple of things:

• The data on the plant floor comes from multiple sources via different interfaces.
• Some data is used by the edge device, and some is passed on to cloud applications.
• Some results of cloud applications are sent back to the edge device when needed.
• The interface between the edge device and the cloud can be broken without affecting production.

Challenges and solutions

This all sounds wonderful! However, there are some challenges to this approach. Security, latency, reliability, and connectivity cannot be guaranteed. These shortcomings can be minimized through the use of quality hardware and software, redundant cloud connections, and network monitoring software.

And of course, cost is always a consideration. Cloud services are not free, and typically have higher costs the more data is transferred, and the more data that is stored.

If you want to learn more about the applications of cloud and edge computing in the IIoT world, check out our course, IoT Basics: Introduction to Industrial Applications.

Future trends and innovations

The use of AI and machine learning is on the rise, and each will prove to be invaluable to cloud applications in better understanding and improving operations, from detecting negative trends to providing preventative maintenance suggestions and allowing for ease of scaling the process.

5G connectivity is a newer, robust, fast communication protocol that will help integrate wireless sensors, devices, and data transport into the plant data infrastructure.

Edge-to-edge computing uses resources at the edge instead of at the core of the network. In other words, edge devices in the plant or between facilities can talk with each other to simplify data paths and provide data where and when it is needed.

When the manufacturing site uses multiple edge devices to provide data and resources across production lines, it is referred to as fog computing. From the perspective of the cloud, data is provided to cloud applications from multiple sources, making it seem like a fog of data below the cloud.

Conclusion

In this article, I have presented Edge and Cloud computing in simple terms. I hope you now understand these concepts in practical terms and are able to see the importance of Edge and Cloud computing in driving automation innovation in IIoT.

These developments have a very high potential for a positive impact on future industrial processes by providing data where it is needed, when it is needed, with greater consistency and granularity than ever before.

As mentioned before, we have several courses on Industrial IoT on our platform. If you want to improve your skills, make sure to check out realpars.com/courses. Under Skill Paths, filter for Industrial IoT.

If you're a plant manager or in a similar role and think your team could benefit from Industrial IoT training, check out realpars.com/business. Add your contact info, and our team will quickly get in touch to see how we can help your team develop their skills.

And if you want to learn more, here’s a list of the courses we mentioned in this blog post:

Course #1: Internet of Things (IoT): Learn the Basics

Course #2: IIoT Fundamentals: Smart Sensors & Actuators in Automation

Course #3: IoT Basics: Introduction to Industrial Applications

‍