Creating Modern, Open Enterprise Architectures in the IIoT Age – Part One

The concept of an open enterprise architecture that links plant floor operations with business operations across an entire corporate entity has been around for a while in many industrial sectors.  But making this concept a reality remains challenging.  This is particularly true for those companies that lack huge IT staffs or budgets.   IIoT, with its compelling promise of accessing, aggregating, and analyzing data from previously stranded assets and systems to improve decision support and thus business performance, represents a further disruption.

Before terms such as “Industrie 4.0” or “IIoT” entered the popular vernacular, most industrial organizations had different data, hardware, and software at each of their plants, with no elegant way to get plant data to the corporate level.

As IIoT awareness spread, demand has grown at the corporate level to acquire, view, and analyze more data from the plant operations and turn it into actionable information. This could save companies millions of dollars, improve decision making, enable centralized management, and pave the way for new technologies like machine learning and predictive analytics. However, successfully transitioning to an enterprise system architecture is no small task. To meet these new demands, industrial organizations need an automated process for delivering plant information to the corporate level in an accurate, standardized, efficient, and secure fashion.

Consider the Whole Enterprise

For most organizations, the first step toward an effective enterprise architecture will be to shift away from thinking of operational technologies (OT) and information technologies (IT) as separate worlds. Instead, they should align OT and IT so that operational data can be shared effectively with business applications.

Also, the challenges of building an enterprise architecture shouldn’t be addressed with top-down thinking. Although it may sound paradoxical, the demand for a centralized enterprise architecture comes from the top, but should be built from the bottom up, starting right at the sensor level in the plant, and always with an eye toward business objectives.

Furthermore, companies should stop thinking of each plant as an island and, instead, look at it as part of the larger corporate system with common standards and data transport mechanisms. Companies must ask themselves: What do we need to do at the plant level to support the enterprise?  This almost always requires secure connectivity.

Organizations need to work on balancing the basic need for security with the need for data. Fortunately, existing technologies can be used to get data up to the central system in an open format without compromising operations. While “air gapping” was often employed in the past as a security measure, this approach isolates OT from IT and thus won’t serve the enterprise’s data needs going forward.  To ensure proper security, data should be encrypted when shared across sites. For those plants that use PLCs, it’s also often helpful to place an edge gateway next to each PLC, so you can get PLC data into an open format while keeping it secure.

Use Open Standards

To build a system with central visualization and administration, it is essential to start developing standards across the entire enterprise. To bring different plants that have different PLCs, tags, addressing schemes, and so on, into a standardized system, it’s often helpful to employ an open source protocol, supported by many applications.

Some common open standards are: OPC UA and MQTT, both primarily used to get data from devices; SQL for working with SQL databases; and APIs (either OPC UA, SOAP or REST) for integrating with other systems.

Once your organization chooses an open standard, you can then standardize your data models so data from all plants will look the same when sent up to corporate.  This is often a more practical approach than gathering different looking data from different sites and translating it at the top.

The protocol you choose as your model should have built-in encryption, as well as “stateful awareness,” in other words, tells you if you’re connected to a specific device. This combination of encryption and stateful awareness provides one approach for getting operational data to the business side in a standardized way without compromising operational security. OPC UA and MQTT protocols both offer built-in security and stateful awareness.

Part Two, which is scheduled for Industrial IoT/Industrie 4.0 Viewpoints Tuesday, October 17th, will discuss "Why Open Architectures", "Technology Migration and Integration", and conclude with "Centralized System Administration Needed" .