What is the Edge in Industrial IoT?

The first thing to do is to lay the groundwork and clarify what we mean by «edge». It is worth noting that the term «edge» is not equally understood by all industries (this is especially notorious in the case of the telecommunication industry, where the «edge» is literally the «edge» of the network, i.e. a node of the network itself).

In Industrial IoT however, e most wideskingpread consideration and the one we use at Barbara IoT is that, the «edge» is the level closest to the physical world. It is the «T» in IoT: the devices. This includes both the set of sensors and actuators that interact with the physical world, as well as the gateways, hubs and other IoT nodes that communicate locally with the former.

The edge is the opposite to the cloud. While the cloud represents a set of services and systems that are remote and far from where the data is captured, the edge is the opposite; it is the local layer and the set of elements close to things and data collection.

Edge Computing and Automation

The term «edge computing» refers to the set of techniques directed to process, analyse and exploit the data collected through all the elements that are part of the edge. Edge Computing is a distributed IT architecture that moves computing resources as close as possible to the source of the data, as opposed to processing data in the cloud or in data centers. As opposed to cloud computing, the execution of algorithms and decision-making is performed at the edg without the need to send all the information to the Cloud. It moves all the computing power from the Cloud to the Edge from a central to a decentralized data exploitation model. Edge computing enables data generated by the Internet of things ( IoT) to be processed close to its source instead of being sent over long distances to the data center or the cloud. Edge computing is a growing trend because it decreases the latency, bandwith and overhead of centralized data centers by moving the device workload closer to the user.

‍

The relevancy of Edge Computing is not only because of the features mentioned above, but  because of data mining possibilities that Edge Computing Technology offers.

Now Artificial Intelligence and Machine Learning technologies can be greatly exploited from a distributed intelligence model.

Artificial Intelligence technologies are based on the execution of multiple complex algorithms that allow machines to «think», and to make decisions without the need for human intervention, simply from the data they process. And many of these algorithms can be run now close to the source of that data, i.e at the edge. The value of IoT Edge is predicated on the ability to capture parameter and event data and process it into insights.  IoT is one of the main avenues for collecting the data artificial intelligence algorithms need. The growing use of these technologies is by no means, a major boost for the use of Edge Computing with IoT. Edge computing is a top priority for organizations looking for effective ways to modernize operations and implement virtualization. The manufacturing industry is making a shift towards merging information technology (IT) with operational technology (OT) for more transparency, improved efficiency, and more timely data analysis. Manufacturers need to reduce plant emissions, create richer customer experiences, support resilient supplychains, as well as minimize downtime, and detect problems before they impact production.

‍

Recommended reading: Interoperability in industry: Why is it essential to digitalize the sector? ‍

AI and ML at the Edge can drive automation

Edge has advantages in many verticals, industrial automation, energy grid automation, water management automation. Automation is currently the number one driver created or enabled by AI.

AI and ML are valuable for detecting anomalies: data can be collected from machines that are on the shop floor, looking at patterns and training the machine learning model for patterns that would indicate that a machine is performing in an unusual  way.

Edge computing allows manufacturers to implement automation across factory floor and supply chain processes through advanced robotics and machine-to-machine communication closer to the source, rather than sending data to a server for analysis andresponse.

Gathering,analyzing, and acting on data on the field in real-time offers profound benefits. Reducing downtime, accurately predicting maintenance, and improving overall product quality results in higher yield, reduced waste,  and lower overall costs.

For Industry, the intelligence is increasingly coming from the deployment of machine learning (ML) and artificial intelligence (AI) technologies at the edge .

Acciona success story‍

The great challenge from the technological perspective was based on the connectivity of a protocol as specific as OPC-UA. It was essential to be able to perform a secure and guaranteed authentication to allow it to connect to the server in order to read parameters and in isolation from the operation, so that it would not affect the normal operation of the plant.

Considering the different types of hardware available, both connectivity and sensors, there was a need for a flexible and cybersecure technology that would allow connectivity with multiple devices while ensuring security of the devices and the extracted data.

ACCIONA managed to reduce in 250.000 € per plant with a better usage of chemicals by deploying AI at the Edge. Download the success story here‍

‍Cybersecurity the main challenge for IoT Edge

The set of elements that make up the edge is by far the most vulnerable point of the entire IoT cybersecurity chain. And the main reason is the lack of firmware updates.

As users in mature sectors such as personal computers and cell phones, we are more than used to receiving notifications of new versions, security patches, etc. However, in the IoT world this is far from the norm. There are mainly two reasons why IoT devices on the edge are not being updated in the same way that our phones and computers are:

1. The immaturity of the industrial IoT market means that companies are focusing in other areas of Industrial IoT rather than in cybersecurity

2. Managing a distributed, remote and highly heterogeneous environment can be very complex without the right tools.

‍

Edge Computing is becoming part of the business value for industries that seek to leverage real time data of their operations. To do so you need the capabilities of a technology that is able to securely capture data regardless the type of sources and equipments and able to execute Edge Computing.

‍

At Barbara we connect, deploy and scale Edge Apps and enable cybsersecure intelligence at the edge. If you want to know how Barbara can assist you in this journey, do contact us.

‍