Digital technology on the factory floor is nothing new: Many manufacturers already incorporate automation and sensors into their production lines. But a broad move is under way to integrate these types of operational technologies (OT) with enterprise IT.
When OT and IT work together as part of the Internet of Things (IoT), information from the factory floor can be leveraged by managers and executives across the business to make better decisions about processes, resources and timetables. The resulting benefits can include lower costs, higher margins, greater agility and a more competitively differentiated value proposition to the customer.
Reengineering the Network
As part of the integration of OT and IT, factory floor controls and sensors are connected to the enterprise IP network so they can send data upstream to databases, analytic applications and dashboards.
This integration requires the right infrastructure, including:
- Ruggedized network devices: Conditions on the factory floor often require network devices that can stand up to greater temperature extremes and more dust than is found in a data center or wiring closet. In some cases, it may be necessary to protect these devices from moisture and vibration as well.
- Appropriate network segmentation: Existing OT networks are typically linked on a single, unsegmented network. That can be problematic in an IP network environment because industrial sensors tend to be very chatty. It is usually a good idea to segment sensors into separate broadcast domains.
- New security controls: OT networks have historically been “air gapped” from the rest of the world. Once they are connected to the enterprise IP network — which is, of course, connected to the Internet — they become vulnerable to outside threats and must be secured accordingly.
Turning Data into Actionable Insights
From a data perspective, manufacturers also have to get a few things right:
- Data conversion: Industrial devices use data protocols such as Modbus, EtherNetIP and PROFINET. To integrate OT data into enterprise applications, these protocols have to be converted into more mainstream formats such as XML.
- Data aggregation and analytics: Once data is converted, it has to be aggregated, normalized and otherwise processed so it can be coherently analyzed to reveal actionable real-time and historical insights.
- Delivery and visualization: Analytic results alone are insufficient to drive action. For that to happen, results have to be intelligibly presented to the right business decision-makers at the right time. This is best done through graphically intuitive dashboards, threshold-based alerts and other data presentation techniques.
All of these steps should be driven by the manufacturer’s particular requirements and objectives. A high-volume producer of low-cost consumer goods will obviously take a different approach to IoT than a company that manufactures precision components for industrial clients.
Every manufacturer, however, can benefit substantially by leveraging the wealth of insights generated by a connected OT infrastructure. In fact, IoT integration is ultimately going to become a competitive requirement for every manufacturer in every market segment.