Manufacturing facilities generate massive amounts of operational data from their automated production equipment, condition monitoring devices, and other sensors and systems. As companies are becoming more aware of the potential marooned in these assets, they are asking how Industrial Internet of Things (IIoT) initiatives can help tap into this information and create useful insights. But many attempts to tackle this via enterprise-wide mega-projects fail to meet expectations because of the sheer size and complexity. Perhaps a better approach is to begin with “little” data at the source to build up to big data using edge computing, focused applications, and open connectivity.
Digital transformation is never a one-and-done task. But taking on a project that is too big can quickly derail the efforts. Just connecting the many different data producers can be hard, yet effectively transmitting, processing, and storing that data — on premise or in the cloud — is an equally massive undertaking. Therefore, the most successful projects start from the bottom up employing machine-level data instead of from the top down at the enterprise level ( Figure 1).
Focusing the field of view on specific challenges and assets can deliver immediate returns compared with gathering all possible data and trying to discover possibilities without a clear problem statement. Plant personnel can prioritize insights that address everyday operational issues and select the most relevant data from existing or even new sensing points. But that uncovers a new challenge: the data of interest is usually not actively being stored in the control system.
For instance, when experiencing excessive downtime with a new machine, not collecting data means the operator cannot find the root cause. Every action taken is reactive to observations after the failure occurs. Applying a little-data project, the user can gather data from the control system along with new air pressure and vibration sensors. By collecting and analyzing data in real time within the context of the machine, the operators can quickly find the root cause and address the issues, leading to improved uptime.
Production data sources are generally part of the operational technology (OT) domain, which includes programmable logic controllers (PLCs), motion devices, and many types of sensors and instruments. But communicating, storing, and processing massive amounts of data requires information technology (IT) capabilities.
The key to bridging OT and IT— and enabling personnel formerly on both sides of this divide to work collaboratively — is found in a new generation of edge controllers (Figure 2). Edge controllers combine real-time deterministic control utilizing IEC 61131-3 languages with general-purpose Linux-based computing to create a powerful IIoT platform that can be built into new designs or incorporated into legacy systems.
This new generation of controller also brings together traditional OT protocols like PROFINET and Modbus TCP with modern IT protocols like OPC UA and MQTT, thus allowing seamless connectivity between edge data sources and higher-level enterprise platforms such as MES, ERP, maintenance management, and other analytics systems.
Many end users recognize they can benefit from IIoT but may be rightfully concerned that a large-scale assault on big data may not succeed. Instead, by approaching IIoT with little data, edge controllers, and targeted analytics, these users will achieve early returns, helping to propel them along faster on their digital transformation journey.
This article was written by Derek Thomas, Vice President of Marketing and Strategy for Emerson’s Machine Automation Solutions business, St. Louis, MO. For more information, visit here .