Whatever the expression—IoT, IIoT (Industrial IoT), IoE (Everything), Industrial Internet—there’s a monolithic sense to its use, as if it denotes a sudden and universal departure from past practices. But this isn’t usually the case in most applications which we’ll refer to as IIoT, as new deployments invariably augment existing IT and OT systems rather than replace them.
“Increased data visibility, aggregation and context—all provided by the deployment of new IIoT technologies layered upon existing IT and OT architectures”
A walk around any manufacturing operation proves this with the geologic layers of technology eras clear to see, and IIoT implementations will be no different. Even if the words are changed to the more generic “digital disruption,” used by analysts frustrated by the “internet” and “things” words in IIoT, we should not expect a moment of disruption—but instead a steady progress of new products and services enabling organizations to more cost-effectively generate, collect, integrate and analyze sensor data.
What this means is the IoT poster examples of smart cities, brilliant machines, and distributed operations networks will be the greenfield exceptions for IIoT implementations. Despite the estimates of trillions of dollars of benefit, ask any analyst, and expected millions of new sensors and wireless connections, ask any vendor, there is simply an overwhelming investment in existing capital assets that needs to be incorporated in the IIoT vision.
So if the networks, systems and sensors aren’t going to be ripped out and replaced with new smart sensors, wireless networks and artificial intelligence bots seeking insight from aggregated big data, what is the IIoT future we should expect for existing brownfield operations? This future vision should start with the required benefits to the business from any new investment.
First, visibility to data will need to increase beyond the real-time control network environment. Second, new data points will often have to be added, in many cases with wireless sensors. And third, more contextual relationships will need to be created among a larger number of related data sources.
Taken together, these improvements in data visibility, new sensors and aggregation will provide the next level of business performance improvements in plant operations. The result will be better product quality, improved yields, higher margins and enhanced safety metrics.
What will not change, often to the chagrin and dismay of some vendors, is the basic architecture for data generation, collection and storage. The pioneering work in manufacturing control models and systems—which will become the sensor, connectivity, control systems and data flows of the IIoT—will be familiar to customers because it’s the same model they’ve leveraged for the last 20 or 30 years. The new capabilities in data collection through wireless sensors and networks, combined with insights from data analytics, will, however, represent new and better ways to accomplish operational improvement tasks.
With this framework established for an incremental IIoT approach to data visibility, new sensors and aggregation in existing plants—here are three scenarios for brownfield industrial operations which depict how the IIoT might layer into existing IT and OT system architectures.
Upstream is an oil and gas industry term for exploration and extraction of petroleum, but it’s also useful as a term for looking back from any production environment to its supply chain of ingredients, feed stocks and raw materials. The location and condition of these materials are of course critical to the production process, and integrating data preceding their actual use can improve production outcomes.
An example from the consumer packaged goods industry is tracking ingredients in terms of aging and temperature exposure in railcars while waiting for use in plant. This could be easily accomplished with wireless sensors deployed on the railcars upon arrival, with collected data stored in a cloud-based sensor network.
The data from these sensors provides new points of measurement which are now visible to the operator or engineer because they are seeing beyond their control network. This data can be viewed along with the manufacturing process data to add context and aid understanding of the dynamics of ingredient production, transportation and exposure on final product quality and yield. Another benefit is added information derived through the ability to accurately understand and compute the causes of product variability from the ingredient data.
Keeping with the oil and gas industry terminology for production process, the next example doesn’t reflect an expansion in data visibility forwards or backwards, but instead focuses on the plant itself by adding depth to the existing data for increased insight.
The example here is applicable to any number of discrete and process manufacturing plants with existing plant architectures and control models which don’t provide the desired visibility into plant operations. This happens both in remote management scenarios where sensors replace the eyes and ears of the engineers who are now offsite, as well as simple onsite configurations.
For example, an exhaust heat recovery system might improve performance of a natural gas turbine, but would also require monitoring of both the system and the turbine to measure and optimize overall operation. New wireless sensors could be added to complement existing real-time control data. The resulting triad of visibility to new data, data aggregation and context among data sets can provide improved operational results.
Finally, downstream visibility to data is also an area for IIoT implementation, for example with remote monitoring of assets such as water/wastewater pump stations, natural gas pipeline compressor stations, and electrical substations. End users can contract with vendors that will provide expert services for monitoring machine health and supplying advice regarding recommended operating improvements and predictive maintenance.
Therefore the visibility benefit in this case is reversed; the end customer is providing, instead of gathering, data visibility to the vendor in exchange for the added value delivered by the vendor’s expertise. The aggregation benefit is available to both the end customer and the vendor.
The end customer can aggregate remote asset data with data from process, MES, AMS and other systems to get an improved view of their operations. For the vendor, they now have the ability to integrate asset production data with customer data by industry and location, with actual sensor data from the asset, and probably maintenance data.
This is for many the leading case for the IIoT model, the heart of the “smart, connected, asset” or “servicization” model , as it provides access to a lifetime arc of asset operation and maintenance. The result would be better products and improved prescriptive guidance to customers on the right asset models and operating guidelines based on observed experience.
Each of these scenarios provides existing operations with increased data visibility, aggregation and context—all provided by the deployment of new IIoT technologies layered upon existing IT and OT architectures. Many deployments will include connectivity to new sensors, and often cloud-based storage for aggregating and providing access to new data.