Representative End User Clients
Representative Automation Clients
Representative Software Clients
Executive Overview
When digital transformation first emerged as a trend, various suppliers positioned their offerings as “ready for Industrial IoT,” or “ready for Industry 4.0.” Initially, this was largely marketing speak, with few if any sustainable strategies to back it up. However, over time, industrial automation suppliers, machine and equipment builders, and end users have 
developed and started to implement effective digital transformation strategies.
The term “Industrial IoT edge” has also emerged. This recognizes that cloud-based digital transformation strategies require data from and access to the physical devices, assets, machines, processes, and applications that reside on the factory or plant floor. The “edge” was initially viewed as the place where industrial network infrastructure devices like switches, gateways or routers, as well as endpoint devices, connected to the Internet. Internet connectivity and automation protocol conversion were the main tasks.
Since then, the role of the Industrial IoT edge has developed rapidly. Today, edge functionality ranging from data preprocessing to artificial intelligence (AI) is being integrated into a broadening variety of systems and intelligent devices. These include traditional automation devices like PLCs as well as edge servers hosting a local cloud.
At the same time, new methods for configuring and programming Industrial IoT edge devices have emerged. Modern smartphones provide a virtual blueprint for how the firmware and application software in future automation systems can be deployed efficiently to devices and updated from a central server. High-level programming languages and configuration approaches from the IT world are also descending to the edge in the form of container support and hardware and software virtualization.
All these developments represent gamechangers in the automation industry. But end user organizations and machine and equipment builders alike now need software engineers and service experts with skillsets that are quite different from those required for traditional PLC, CNC, or DCS programming and maintenance and more oriented toward IT skills.
Digital Transformation Drives Need for Powerful Edge Computing
Once the protocol converters were in place at the edge to connect factory assets to the Internet, the next evolution was to enable the assets (machines, equipment, controllers, and sensors) to deliver digital data to data centers and enterprise clouds for evaluation and, ultimately, to help improve business processes. A data tsunami resulted, triggering the need to filter the data by smart data mining and filtering at the edge, so only preprocessed or screened data is sent onwards to data centers or clouds. To be able to mine and filter the data close to the plant-floor machines and equipment, edge computing capabilities were required. As a result, edge devices evolved into powerful computers as CPU horsepower and storage capabilities were added to accommodate these requirements.
Many production companies were also concerned about transmitting internal operations data to a remote data center or cloud. This resulted in the adoption of on-premise local clouds and the need for even more powerful edge computers and servers.
Who Owns the Industrial IoT Edge ?
Rapid evolution at the industrial edge has brought powerful computing power to edge systems. Many users realize they now have incremental computer power very close to their machines and equipment. This raises the possibility to also use this compute power for machine control. One avenue is to run a soft PLC in the edge computer while preserving the legacy remote I/O in the machines or process equipment. IT companies like HPE, Cisco, Dell, and others are responding by providing edge computers and servers that meet the requirements of industrial customers.
On the OT side, industrial automation suppliers started introducing edge co-processor modules capable of operating on the backplanes of PLCs and other control devices. This generated new competition with IT device suppliers as industrial automation providers tried to capture a substantial part of the edge compute market on their own – again a gamechanger in automation and the associated business models. Certainly, customers play a major role in the decision as to who will supply them with the edge compute infrastructure: their automation suppliers, IT suppliers, or a combination of the two.
Machine Control at the Industrial IoT Edge ?
Integrating edge computing capabilities with machinery automation raises concern as to its impact on the extremely high reliability of traditional logic, motion, and process control systems. If edge computing functionalities are merged with automation tasks, the risk of shutdowns caused by edge operations needs to be considered and measures taken to maintain the reliability of the traditional automation systems. Some end users request that their machine suppliers keep edge functionalities and automation processes strictly independent from each other to secure ongoing operation even if the edge compute device fails.
Typically, in these cases, should the edge compute functionality drop out, the machine will continue running in the traditional automated mode. For example, in cases where the edge functionality is there to support AI-based quality control, should the edge device fail, the machine will keep operating. The only downside is that during the edge compute drop out, the traditional quality software routines would likely reject more products than they would with the AI-based quality inspection. Consequently, users must carefully evaluate the potential advantages and pitfalls of integrating new edge-enabled functionalities and traditional automation processes in a common edge computer.
Table of Contents
- Executive Overview
- Digital Transformation Drives Need for Powerful Edge Computing
- Who Owns the Industrial IoT Edge?
- AI and ML Drive Compute to the Edge
- ARC Industrial IoT Edge 4.0 Defined
- Tailoring the Framework to Specific Use Cases
- Recommendations
ARC Advisory Group clients can view the complete report at ARC Client Portal
If you would like to buy this report or obtain information about how to become a client, please Contact Us