Integrating IIoT technologies to maximize facility operations

Gain valuable data insights by integrating operations, information technology (IT) systems, and creating a more effective Industrial Internet of Things (IIoT) solution.
By Michael Risse, Seeq Corp. April 9, 2018

Providing subject matter experts with visual representations of data allows them to interact directly to solve problems, without the need for assistance from data scientists and IT personnel. Courtesy: Seeq Corp.Control Engineering (CE) asked Michael Risse, vice president of Seeq, for advice on how controls, automation, and instrumentation help with integration using Industrial Internet of Things (IIoT) technologies. Integrating operations and information technology (IT) systems can be made easier with the following considerations.

CE: When those in automation and operations consider integrating IIoT concepts, what technologies are they most often talking about? Also, what technologies should be considered but may be overlooked?

Risse: The basic components and systems of the IIoT have been a fact of life in industrial automation and operations for decades: sensor, data collection, application, storage, and analytics. What’s needed are insights to improve outcomes in terms of quality, yield, margins, safety, etc.

Most of the promises of digitization transformation, IoT, etc., are about bringing that opportunity for insight to things that didn’t have it (connected cars, smart refrigerators, brilliant parking lots, etc.) to solve the same types of problems seen in industrial plants and facilities by providing predictive analytics, operational insight and visibility to things not in front of them via remote access. The next step is to extend the visible range of results to partners, upstream (supply chain) and downstream (customer use)—and across organizations (x-plants).

The IIoT is not the same old stuff in industrial applications because there are very significant innovations driving opportunities and lower cost such as: 

  • Cloud: connectivity, data collection, sharing, simplicity
  • Sensors: more, cheaper, wireless, edge processing, OPC unified architecture (UA) and message queuing telemetry transport (MQQT) protocol support
  • Analytics: more data can now be used to provide much more insight by implementing new approaches. Software platforms makes this possible by using technologies such as Big Data, machine learning, open source, etc.
  • Business models: remote monitoring of assets by vendors, selling thrust versus engines, distributing risks.

CE: How can controls, automation, and instrumentation help with integration and use of IIoT technologies?

Risse: Stick to a business case, start small and create value. Just don’t overthink this as top down—IIoT projects often take too long and cost too much. Try to find a greenfield in a brownfield scenario—something new in something existing (use case should depend on what matters most—ingredient quality, energy use, emission compliance, etc.) For example: 

  • Skunkworks: Raspberry Pi to track sensor data on wireless to inform/contextualize a sensor.
  • Midsize: cloud IIoT platform for disconnected data sets (remote assets) that expand visible range for operators, or optimization context for engineers.
  • High level: remote monitoring center to centralize operations.

CE: What value is being created and how, with integration of operations and information technology (IT) systems?

Risse: Value doesn’t have to be in dollars. It could be safety or any other priority for the organization such as regulatory compliance. The value is in the time to insight to improvement. Or, to put it another way: sooner means value, and the insight means value, and the sooner the insight is the square of value (or multiplier) because it’s sooner (like area under the curve). How it’s created is insight.

Many vendors love to talk about sensors, wireless, and other technologies, but the point of all this is improved outcomes through quicker insights and delivering the profit impact of doing something better, sooner. Also, the value could be the same insight an organization has been wanting to achieve for years. Now it’s profitable to do so because the cost of achieving the insight went down (cheaper data storage, collection, new analytics, faster insights). The IIoT is a point in progression in terms of lower costs for most of the components in these types of systems.

CE: How can data analytics help make sense of existing data and the additional information created when previously disparate systems are connected?

Risse: Data analytics is a huge help because one of the keys for successful IIoT implementations is tapping the innovation in analytics technologies so subject matter experts (SMEs) can find insights faster. This allows them to answer the questions they have wanted to analyze but were just too hard/too long to do with traditional tools such as spreadsheets.

So for existing data, there is untapped potential in data already collected no matter where it resides. Data analytics can enable a bridge across data silos to simplify contextualization by putting data in its frame of reference. Previously this was done mostly by hand, mostly by an SME in Microsoft Excel or as part of a massive information technology (IT) project. Now this can be done by any SME. With new data sources, this will become even more of an issue. Perhaps an answer is data lakes or data roll ups for multi-plant scenarios, but data collected in one place without analytics is just a bigger headache than when it’s spread out.

CE: What other advice or tips would you offer on IIoT integration relevant for Control Engineering subscribers?

Risse: Beware of all systems requiring new data storage. Typically, it’s not the data that needs help; it’s the analysis of the data to produce insights. Starting small and soon should not require undertaking some Big Data transformation/cloud movement effort.

There has to be a midpoint between hopelessly global (IT top down takes forever) and mindlessly local (plant-by-plant skunkworks with disconnected and nonstandard tools or approaches). This midpoint can be implemented by getting started now on existing problems. Even if one tosses out a local project due to lack of success, it can inform what you need to know and how you can make better decisions, and at very quickly and with low implementation costs. If there are no local initiatives, competitors will outperform laggards; so it’s best to get started sooner rather than later.

Michael Risse is vice president at Seeq Corp. Edited by Emily Guenther, associate content manager, Control Engineering, CFE Media,


KEYWORDS: Industrial Internet of Things (IIoT), Big Data

Exploring IIoT integration best practices

How to use data analytics for a successful IIoT integration

The value behind integrating systems and operations.

Consider this

How will integrating IIoT technologies help maximize efficiency, safety, and operations?