Industrial Analytics: Predictive and Prescriptive

A few blogs back we looked at growing interest in extracting value from IoT data through industrial analytics.  This interest has not sprouted up overnight.  Since the beginning of computer-assisted control systems, plant engineers and managers have been using their increasingly powerful and sophisticated tools to gather data, and then use the data to improve their processes.

For much of that time, the idea was to collect data in a database, and then at the end of the month or quarter, run various analytical tools on the data to see where the problems and bottlenecks were, and what could be changed.  This approach had some value, but it is essentially a reactive model.  Today, there is a general trend underway to go beyond simple reaction like this, and move towards the ability to predict problems, and if possible prescribe a solution.  In a recent blog, Blurred Lines Between Predictive and Prescriptive Analytics Mike Guilfoyle at ARC Advisory Group explains the value of each of these approaches to analytics, as well as their differences.

He breaks down this kind of pro-active analysis into three parts: performance, predictive and prescriptive, distinguished as follows:

  • Performance describes what is happening or has happened, and is the starting point of all analytics, reactive or pro-active.  The focus here is on current or past performance.
  • Predictive looks forward to what is most probable to happen, given the current conditions, using Big Data, machine learning, and other IT tools.
  • Prescriptive uses all of the above inputs, and adds to that a knowledge base and decision-making algorithms to prescribe what action can or should be taken.  In some instances, the system might actually even carry out the action, which is referred to as “closed-loop control.”

Guilfoyle goes on to identify some of important differences between predictive and presciptive analytics.  In fact, he will be leading a session on analytics best practices at the ARC Industry Forum this week.  You may not be able to attend, but his article is a good introduction.

In any case, the trend towards predictive and prescriptive analytics and any kind of closed-loop control based on such approaches highlights the need for secure, real-time access to plant data.  It is yet another example of the closing gap between OT and IT, and is an unmistakable benefit of the Industrial IoT.

Will this be the year that your enterprise makes the IIoT leap?

For the second January in a row, we’re using this lead issue to look ahead at the industry conversations likely to follow over the next 11 months. Like last year, there’s still no bigger buzz than the impact that digital transformation and the industrial internet is having both on work and on the people who do work.

I’m writing this note on the 10-year anniversary of the launch of the iPhone, which marks a genuine milestone in the history of both internet-enabled communications and mobile computing. As the iPhone evolved and the iPad emerged, savvier organizations and IT workers caught on early to the opportunities available to digitize operations. For example, a close friend who works in commercial real estate directed his teams early on to rethink his organization’s processes as each new Apple device launched, reducing business friction in the field and moving toward nearly paperless operations.

Many other contributors this month round out the digital conversation:

  • IFS CTO Dan Matthews identifies three myths that cause organizations to hesitate on IoT projects.
  • Skkynet’s Bob McIlvride examines how to combine in-house skills with outside expertise to build systems that enable deeper data-driven insights into your assets.
  • Bruce Hawkins and Scott Bruni review the foundational IIoT steps that plant teams can take, noting that roughly 60% of the instrumentation needed for critical assets often already exists.
  • Tech Toolbox’s Sheila Kennedy surveys the network security solutions landscape in an age of IT-OT convergence.
  • Jeff Shiver of People and Processes outlines six steps that can improve the speed and quality of cultural change in your organization.
    Finally, in her Big Picture Interview, Bentley Systems’ Anne-Marie Walters looks ahead to the role that 3D modeling will play in the internet-enabled asset management landscape.

Industrial Analytics: Extracting Value from IoT Data

“Analytics is to data what refining is to oil: The process that turns the resource into a valuable product,” says the opening paragraph of a new survey report, Industrial Analytics Report 2016/17, initiated and governed by the Digital Analytics Association e.V. Germany (DAAG). The report provides a good overview of how executives in Europe and around the world, representing leading manufacturers, system integrators, automation tool vendors, and other institutions, view the value of IIoT analytics, and how this new application space will continue to expand.

The rapid growth of the Industrial Internet of Things (IIoT) is already precipitating a deluge of data, and manufacturers are anticipating much more to come. As they experience this mounting wave, they also recognize the need to extract value from it. Thus, a majority of respondents to the survey said that industrial analytics will become crucially important over the next five years. That value will be due, they believe, to increased revenue from the data sources that the IIoT will tap. The way they see it, analysis of IIoT data will open opportunities for predictive and prescriptive maintenance, better analysis of customers and markets, and a better understanding of how products are actually used in the field.

Most responses indicated that to take full advantage of the data stream, the quality of these analytics will need to gain in sophistication. For example, the majority foresee exchanging spreadsheets for Business Intelligence and advanced analytical tools. These real-time analytical tools are expected to help them evolve from a current ability to merely describe problems towards the capacity to predict the problems, and even prescribe solutions.


Of course, there are challenges to be met. All of this will come at a cost, replied those surveyed, with the largest expenses expected to be for the software and applications needed to gain access to the data and aggregate it. Another challenge is a skills and technology gap in the area of the IIoT infrastructure. In general, a full 78% of the participants rated “interoperability between different system components” as challenging or very challenging. About 60% said the same for “data accuracy,” and about 50% rated “integration with enterprise systems” at that same level of difficulty.

These survey results validate Skkynet’s approach to the IIoT. We believe that companies should not have to get drawn into infrastructure development to reap the benefits of sophisticated analysis of live and historical IIoT data. We provide interoperability through secure, real-time data exchange between remote devices, shop-floor equipment, multiple facilities, and main-office IT departments. Companies accessing our SkkyHub™ service can gain the full value of the IIoT with no development costs or capital expenditure.

Any company looking into IIoT-based industrial analytics should dream big, sharpen their analytical skills, and choose good tools. When they are ready to connect to their data sources, integrate them, and put the results into their analytical systems, they should come to us.