At LNS Research we have been writing a lot about what the Industrial Internet of Things (IIoT) has done to manufacturing operations management (MOM). We have also just tackled the topic of how the IIoT and Cloud technology is changing the way companies look at enterprise resource planning (ERP) systems, as well. In both cases we see the many-to-many communication paths the IIoT offers; changing the way users envision their Operational Architecture. They no longer see systems as hierarchically stacked with structured data flows between layers. Admitted, they do have security concerns, and rightfully so.
The advantages of getting the right information to the right person or system at the right time offers so many possibilities to drive Operational Excellence, especially when you add Cloud platforms and mobility solutions into the mix, that companies pursuing Digital Transformation (DX) are finally embracing this new model. It is our belief at LNS Research that another critical part of manufacturing is going to also get radically impacted by the IIoT and Cloud technology as well; the control layer, especially distributed control systems (DCS’s). But that begs the question:
How distributed is a DCS in reality?
Ever since the introduction of the Honeywell TDC 2000 and the Yokogawa CENTUM control systems in 1975 the standard for process control in continuous (and to a lesser degree batch) process industries has been the DCS. The term DCS chosen by those companies because they were able to move controllers out of a single mainframe style computer in a data center environment to microcomputers and microprocessors throughout the plant in remotely located control rooms. So, while there was a degree of distribution in the DCS’s of that era the reality was that the distribution was very hierarchical and still very much like a data center environment. With the introduction of the Foxboro (now part of Schneider Electric) I/A Series in 1987, more IT in the form of Unix and Ethernet networking moved into the DCS world. Additionally, the hardware was more hardened and could be more distributed, located closer to the process itself and not requiring a control room environment.
During this same era, programmable logic controllers (PLC’s) also emerged as hardened devices capable of being distributed closer to the process in batch and discrete manufacturing. Since that time there has been a blending of PLC’s relay style of control with the DCS’s control loop structure so hybrid devices are now available that work in mixed manufacturing (process & discrete) environments. But there is a commonality in both that is hard to ignore – they really are distributed controller systems, not distributed control systems.
How the IIoT Will Enable Truly Distributed Control
A big part of the promise of the IIoT is the many-to-many connectivity of IIoT enabled devices. As the major automation providers start to roll out their IIoT platforms and solutions, we expect to see them heavily leverage both Cloud technology and edge computing capabilities. Since the security of the Cloud platforms that are targeted at manufacturing, like GE’s Predix, and Azure, IoT Hub is now industrial grade; we expect to see companies start to rethink their control platform strategies.
Recent conversations with companies like Schneider Electric and Emerson lead us to believe that those suppliers that have depth in their sensor portfolio are going to begin to embed capabilities in those devices so that a separate controller becomes redundant. We are already at the point that the actual control loop functionality can be and sometimes is distributed among the devices, but with the IIoT and Cloud, access to operational data such as production value (for setpoint determination) no longer needs to be cascaded down from the ERP to the MES and then to the DCS or PLC. Instead, the Loop devices needing a setpoint can get it from the original source directly. Edge intelligence is what makes this viable in that the rationality of the decisions being made at the device level can be validated and, with predictive analytics performance, targets can be monitored and managed. A properly defined asset hierarchy will be the key element in ensuring that information gets to the right devices.
So, it is our belief that we are on the verge of seeing truly distributed control instead of just controllers. The suppliers that make the move first will gain the support of the manufacturing pioneers that are ready to pilot IIoT technology and make it part of their Digital Transformation journey today. We see evidence this evolution is beginning. ExxonMobil has engaged Lockheed Martin IT Consulting to help them design a next generation of control systems and at Cisco Live in Las Vegas in July Cisco announced edge devices that will enable the migration towards this objective.
This NEW eBook explains the Industrial Internet of Things (IIoT), Cloud, Big Data Analytics, and Mobility, as well as how these technology trends are enabling innovative approaches to communication, asset maintenance, and products and services that drive enormous business value.