Getting the facts out of your factory

Article

Getting the facts out of your factory

Sharing information between IT and OT systems

Within manufacturing, a distinction is made between IT, which deals with all forms of technology that handle information, and OT (Operational Technology), which consists of the hardware and software that drive production facilities. Historically, IT and OT are barely connected, but there is an increasing need for sharing of information between the two. How can you connect them?

Getting the facts out of your factory


Sharing information between IT and OT systems

Within manufacturing, a distinction is made between IT, which deals with all types of technology that handle information, and OT (Operational Technology), which consists of the hardware and software that drive production facilities. OT keeps robots moving, factory lines producing and energy flowing, whereas IT supports the business in other ways such as inventory management, resource planning and accounting.
These are the systems that form the backbone of every form of automated production, whether we are talking about oil refineries, car factories or power plants. Historically, IT and OT are not connected at all, or the connection is made for very specific purposes (e.g. through a manufacturing execution system [MES] to track the conversion from raw materials into products). However, there is an increasing need for sharing information between both systems. For example, think about connecting sensors that measure wear (OT) to maintenance planning tools (IT) or connecting real-time energy price information (IT) to the factory control system (OT).
Of course, the existing IT and OT systems within an organization might not include all data needed for specific Industry 4.0 applications. Additional sensors could be installed outside of the OT systems, directly connected to the internet. Starting with OT integration makes sense though, as this leverages the sensors and machine networks already in place and often allows access to historical data (e.g. through an existing MES). But how can you connect IT and OT with each other?


Transitioning to the cloud

The business case for moving an organization’s IT system into the cloud has been strong for some time and most, if not all, organizations have moved some activities to SaaS (Software as a Service) or other forms of cloud solutions (e.g. PaaS, IaaS). Benefits include flexible scalability, seamless upgrades and data available everywhere.
Currently, we are in a transition period, in which some applications have been migrated into the cloud and others still run on-premise. This results in a hybrid IT landscape, which will gradually evolve into a fully cloud-based IT landscape. However, in a production environment, we do not only have to deal with IT systems, but also with OT systems. These systems detect changes using sensors and cause changes using actuators like motors or valves.
Such control systems, currently embedded in PLCs and industrial computers, cannot (easily) be migrated to the cloud. The response times (typically below 1 millisecond) associated with the needed real-time computing simply cannot be guaranteed by cloud solutions: if we assume signals travel with the speed of light, data packages are not delayed by switches and routers and the servers require no processing time at all, “the cloud” could never be further away than 150 kilometers. This back-of-the-envelope calculation shows that in practice the OT will probably not leave the plant site.


Integrating OT with IT through cloud-based integration platforms

OT systems process and produce a lot of information, which is also of interest outside of the manufacturing process itself. Extracting this information from the OT systems and into IT systems (“physical-to-digital” dataflow) and adapting the OT systems based on the gained insight (“digital-to-physical” dataflow) is at the core of the Industry 4.0 revolution. The boundary between OT and IT systems fades, as sensor data and production settings are aggregated in data analytics platforms and are linked to traditional IT systems, such as ERP, production planning systems, and so on.
Meanwhile, the real-time control logic must remain physically close to the production systems to ensure timely response to sensor input and ensure safe operation. What results is a hybrid structure in which traditional IT systems reside in the cloud and the data generating OT systems remain on-premise. The exact position of the boundary between on-premise and cloud should be chosen depending on the specific needs and current infrastructure in place. In general, it would make sense to stick to the same strategy as used in pure IT systems: integrate IT and the data from OT in the cloud to gain the same benefits as moving pure IT solutions into the cloud. A variety of cloud-based integration platforms, specifically designed to handle large amounts of data and utilizing machine learning, exist today and are production proof.


More information on Industry 4.0?

Would you like to know more about Deloitte’s view on Industry 4.0 and how it will affect your business? Please contact André Barneveld Binkhuysen at +31 (0)88 2880003.

Vond u dit nuttig?