[IIC Whitepaper] Industrial Networking Enabling IIoT Communication
The internet is now redefining industries like energy, manufacturing, transportation and healthcare. This new wave is the Industrial Internet of Things (IIoT): an internet of things, machines, computers and people, enabling intelligent industrial operations using advanced data analytics for transformational business outcomes, according to Industrial Internet Consortium Industrial Internet Vocabulary, Version 1.0, May 7, 2015.
The foundation of IIoT is how the industrial assets (the things, machines, sites and environments) can be connected to business professionals and processes. The network is the important infrastructure that supports different application requirements and different deployment situations in the vast range of industry sectors and associated industry-specific applications affected by IIoT (source: The Internet of Things: Mapping the Value Beyond the Hype, McKinsey Global Institute June 2015).
Industrial networking is a collection of technologies at the Internet Protocol (IP) layer and below that enable the transformation of industries. There are many choices in technologies, both existing and emerging. What are the networking needs, what applications to support on the industrial network, and what is the deployment situation and conditions? These are key questions to answer when defining a strategy for technology selection and reaching a concrete deployment solution.
Enabling IIoT Communication
Industrial networking is different from networking for the enterprise or networking for consumers. First there is the convergence of Information Technology (IT) and Operational Technology (OT). Important networking considerations include whether to use wired or wireless, how to support mobility (e.g. vehicles, equipment, robots and workers) and how to reconfigure components. Other factors include the lifecycle of the deployments, physical environmental conditions such as those found in mining and agriculture and electromagnetic conditions where interference from machines and equipment can be a problem. Then we have power. Will it be available? Or will devices need to run on local power such as batteries?
Second there are the technical requirements. They include network latency and jitter, throughput needs, reliability and availability. The requirements can vary from being relaxed to highly demanding. The network must meet the end-to-end performance requirements for applications deployed both at the edge and in the cloud. Service Level Agreements (SLAs) must suit the industrial application’s requirements, which are very different for utility smart metering, agriculture monitoring sensors or the remote operation of a mining drill rig requiring stereoscopic high-quality video feeds and haptic feedback for control.
This white paper sets the stage for a forthcoming publication, the Industrial Internet Networking Framework (IINF), which will complement the Industrial Internet Connectivity Framework by detailing requirements assessing available technologies and providing best practices for how to achieve appropriate IIoT networking solutions.
It introduces a few typical industrial scenarios followed by their impact on networking. The scenarios show that there is no universal or preferred networking technology for IIoT systems and that diverse requirements must be considered when selecting an appropriate solution.
The derived networking requirements lead to the diversity of design considerations, which provides introductory guidance to IIoT solution architects and industrial networking engineers to help them make the right choices.
Future IIoT scenarions and their dependence on Networking
Productivity and profitability depend on keeping the operation running. For example, to maintain the highest level of production in manufacturing, machine maintenance needs to take place during scheduled downtime. This is best when machines can predict when and where faults are about to happen so they can be serviced then. This collected data needs to be communicated across industrial networks for analysis and can also be used for machine learning and for conducting business intelligence.
Inventory control systems also use industrial networks to maintain efficiency and increase inventory throughput. Machine interoperability and network interoperability standards create environments that easily allow industrial operations to be commissioned, managed and integrated into business processes.
Industrial networks improve productivity and performance by communicating essential operational data. For example, condition monitoring and energy monitoring collect data from every piece of equipment in the operation that can be used to create profiles for every piece of equipment, such as power usage across an equipment cycle and vibration profiles during normal operation. This data is often transmitted to the cloud for analysis and, when compared to previous profiles, can contribute to predictive decisions about when to service the equipment, thus pre-empting equipment failure during production.
There are many more emerging scenarios that affect these layers and that need to be supported. See several examples here.
This in an excerpt from the Industrial Networking Enabling IIoT Communication white paper released yesterday by the Industrial Internet Consortium