In a contributed article for Internet of Business, Donna Prlich, chief product officer for Pentaho software at Hitachi Vantara, argues that manufacturing companies must mix IT and OT data to achieve new levels of insight and efficiency.
Until recently, many manufacturers may have wondered whether the post-industrial world would leave any reason (or deliver sufficient profit) for them to exist. Today, technology changes ushered in by the industrial Internet of Things (IIoT) are breathing new life and opportunity into the sector.
However, the extent to which manufacturers will benefit from IIoT will depend on the maturity of their implementations – and to gain the most valuable and most profitable insights, these implementations will have to tap into data from many different sources and blend it in ways that deliver real insight.
In particular, that means bringing together data from two different types of system that have historically been siloed from each other: Information technology (IT) systems, such as enterprise resource planning (ERP) and supply chain management (SCM) software; and operational technology (OT) systems, that are charged with the task of monitoring and/or controlling physical equipment in a manufacturing environment.
Mind the gap
There are good reasons for why IT and OT systems have traditionally been kept separate. Manufacturing and industrial facilities use OT systems mainly to ensure availability. Set up as ‘closed-looped’ systems, their data is disconnected from enterprise IT systems.
Industrial environments, where a small change can trigger a domino effect, adopt systematic, methodical approaches to maintenance.
IT systems, on the other hand, which undergo regular maintenance and upgrades, can afford occasional downtime. After all, they aren’t engineered to handle high-voltage systems or the control rods of a nuclear plant.
Nevertheless, the convergence of IT and OT systems and data is already starting to happen. OT is evolving to work with proven IT technologies, such as security software. By the same token, IT systems are scaling to handle the huge data volumes generated by factory OT systems.
Edge computing – the trend that see data processing and analytics move closer to the machines that generate that data – plays an important role here, as do real-time streaming technologies such as Apache Spark and Apache Kafka, which are enabling companies that adopt IIoT to react more quickly to changes.
The STIWA Group builds bridges
The STIWA Group has a long history in product and high-performance manufacturing automation. Its machines are highly automated, requiring little human intervention to run. The company also provides data about its machines to customers using them and builds the software they need to analyse it – and is using analytics software from Pentaho to automatically process signals and data as a basis for its own product, AMS Analysis-CI.
The processed data includes machine, production and quality data collected by another STIWA product, AMS ZPoint-CI. In other words, this is an example of OT data being explored via a typical IT approach.
This helps the STIWA Group handle tracking for safety-critical products and gain control of complex assembly and manufacturing processes.
A leap of faith needed?
The most common mistake people make when implementing any new technology is solving problems in isolation. This is especially problematic with IIOT, where success depends on big-picture thinking.
Take, for example, a steel factory that wants to improve efficiency by tackling a specific issue that occurs daily: a technician typically looks at the slice of OT data directly relevant to that failure. This could by, say, 20 variables from a supervisory control and data acquisition (SCADA) system.
If, however, the OT data was blended with data produced by environmental control and factory planning systems, that technician might not only solve that specific failure but also be able to prevent future ones from happening. This integration also reveals relationships between components that help to significantly improve overall equipment effectiveness (OEE).
In this case, bridging the divide may have involved a leap of faith, but it’s one that has allowed the company to arrive at valuable new insights.