Bringing IT and OT together: Christine Frank
By J. Tamhane
With IoT becoming more and more of a common topic, the worlds of IT and OT (essentially, the factory) come closer and closer. The AI tools (or to be precise, Machine Learning Algorithms) are presenting numerous opportunities of making sense of the vast data that factories have and can collect. One such dynamic, innovative and humble executive with more than 20 years of experience in industrial automation, process control and BAS systems (safety, controls, industrial networks and fieldbus), is Christine Frank.
For the last eight years, Christine has focused on helping companies bridge IT and OT for results in IIoT, IoT, Industry 4.0 and Smart Manufacturing. She prefers to call IT and OT "as carpet and concrete sides of the house". Indeed, "the concrete side is where the magic happens - manufacturing, process control, building automation etc., and is usually environmentally challenged".
Christine also has quite a valid take on the concept of fail-safe that is all too common in the factories. She points out that "fail-safe happens in hazardous locations like oil & gas as well as machine safety in discrete manufacturing. Fail safe means just that; to fail in a safe state. Industrial safety is meant to keep people and assets from harm, catastrophic events and/or even death". When we compare it to IT, "no one is going to die if they miss an email or a VoIP call. Now, they might feel bad they missed Taco Tuesday but, no harm will come. I think that if we are talking about a cyber-security angle, that is different. Especially, when it comes to medical and financial data", says Christine.
On the topic of regulations, Christine mentions that "OT already has regulatory and compliance standards for electrical controls and process equipment. Where regulations are needed is in cyber security", which spans the entire spectrum of IT and OT. Of course, too much regulation can be harmful to any sector. But, so can a lack of regulation.
There are challenges ahead while bringing in AI into the picture. The most obvious unknown is the cost/benefit of embarking on potentially long projects with hard to define outcomes. "Cost savings that equates to ROI/RoA must far exceed the cost of implementing. If you can't see an ROI or RoA 3X's or more what's the point? Plus, no one wants to be a science experiment or guinea pig in IoT using AI. Therefore, the solution better be solid and help the customer with a competitive edge - i.e. balance sheet", explains Christine.
Christine highlights that "OT has done a terrific job on the cement side of the house with organizations like OPC-UA - connecting all kinds of ICS's, PLC/PAC's, I/O, SCADA, and MES,MOM's that use different protocols, networks and fieldbuses". In the same way, "IT has done a good job of interoperability with BI tools. What is still a problem is the ability to talk the same language from the shop floor to top floor - carpet and concrete", which is where she sees most of the challenges. Indeed, the lack of overall interoperability makes it difficult for Big Data engineering applications and use cases to have speedy returns on investment.
If we are to launch long-term IoT projects that would collect vast amounts of data, we must pay attention to how we intend to use the data. Any value can only come from analysing it. But before we analyse the data, let's bridge the gap between IT and OT and let's truly understand the challenging environment of a factory. As Christine says: "Let's be Frank!".
+49 6033 1817 900
Virginia Str. 24
35510 Butzbach, Germany
on LinkedIn for more showcases.