Caspar Herzberg |
This article first appeared in the Spring 2017 issue of The Record.
A key challenge around Industry 4.0 and the industrial internet of things (IIoT) is creating secure, standardised data and exchanging information between devices, machines and services across various industries.
IEC62541 – the OPC Unified Architecture (OPC UA) standard – is the only mandatory protocol required as part of the criteria checklist of ‘product capabilities for Industry 4.0 products’ created by Germany’s Federal Ministry for Economic Affairs, and Economy and Federal Ministry of Education and Research. This means that essentially, when it comes to adopting Industry 4.0 practices, the OPC UA information model property is a requirement, rather than just an option.
On the face of it, the concept of information modelling seems too complicated for most companies and engineers. In the main, they want to talk about ‘how’ to exchange data and information, rather than ‘what’ it is they are exchanging.
Interestingly, each device and machine builder today provides an information model detailing data and interfaces. We, as humans, have started to adapt to thinking about computers in bits and bytes and hex codes. The new world of devices makes it faster and easier for people to understand what these ‘things’ are and how they work. A service-oriented architecture (SOA) approach in the IT world has existed for a while – but now SOA is moving into the things themselves.
Thinking about connectivity – the ‘how’ I referred to previously – is all well and good, but the real key for IIoT is understanding the ‘what’ – the data and its behaviour, or the ‘things’ themselves.
This is where OPC UA helps as an industrial interoperability framework. Device and machine builders can add and define object-oriented information to their systems and confirm access rights via integrated IT security.
But who is defining this ‘what’? The answer is other associations. These include the AIM organisation for the automatic identification and data capture industry, and Germany’s Mechanical Engineering Industry Association trade groups, which cover everything from injection moulding machines to robotics. Numerous other organisations already define their information in OPC UA servers – a companion of the OPC UA specification.
Supporting this standard does not simply centre around data exchange – a number of additional special services can also be offered in parallel. This means that differentiation inside the device, or outside with external cloud services, is key.
Intelligent devices support multiple parallel information models. This includes the dedicated functionality of the device itself, but also energy data and manufacturing execution system interfaces.
The importance of these industry and cross-industry information models will rise sharply in the future. Manufacturers that do not support these standards could soon be out of business.
Microsoft has deeply integrated OPC UA in its Azure cloud, which allows bi-directional communications with OPC UA enabled devices. This includes device to cloud, which pushes telemetry data to the cloud, as well as cloud to device, which helps manage devices with optional command and control.
Perhaps most importantly, however, is the ease with which Microsoft can understand this data and services. This is key for IIoT, and is exactly why OPC UA is mandatory for Industry 4.0.
Do you still describe the meaning of data in each project? Why not become part of the biggest ecosystem for industrial interoperability, OPC UA, and really use and analyse data instead of just describing it?
Stefan Hoppe is vice president of OPC Foundation