By Joe Perino, Research Analyst, LNS Research   (trans. ServiTecno)

Process industry companies have begun to realize that the data created within their plants is significantly underused.

But often this data does not leave the narrow scope of operations. As a result, many companies are taking steps to gain greater visibility and integrate plant data into broader operational and business analytics.

The expectation is that by combining data relating to production, supply chain, financial management and even data that comes from sources outside the corporate perimeter, greater profitability can be achieved.

Those who are working in this direction are discovering hidden opportunities and new synergies for improvement.

Three must-haves to improve visibility

But what exactly does it mean to have “more visibility”? At LNS Research, we believe that visibility consists of three critical capabilities:

  • Timely access to data in a format and structure that staff can easily analyze
  • Ability to “see what is happening now” by comparing and diagnosing performance within the plant, between plants and throughout the supply chain
  • Ability to "look ahead" with forecasts of future performance to anticipate potential problems

This includes identifying previously unseen patterns, relationships and anomalies that can hold back performance. Together, these three capabilities enable companies to make better decisions and improve the KPIs that are most relevant to business profitability.

Timely access to data

The first must-have of Visibility | Know the data you need: which, when, where and why

Having timely access to data is essential, but unfortunately it is not easy, indeed, it is a real challenge both for the operational technology (OT) and for the IT architecture.

According to recent research, manufacturers are working to evolve their traditional OT architectures by adding pervasive sensors, wireless and wired edge devices, and cloud-based resources.

While it is true that the Plant Historian is the primary source of production data, it must be said that it is not the only source of all the data needed for company operational analyses.

For this purpose, it is necessary to evaluate the inclusion of a data engineer, a figure that is starting to emerge to evaluate what data is needed, where it must be taken from and of course when, taking into account timeliness, the required format and the available sources.

This figure is generally a professional in the production department, even if this is also often a role that is missing in companies. Whether it's an official job or just someone who takes care of it de facto, the data engineer is a key figure to drive the right operational architecture to support visibility.

Diagnosis and performance comparison

The second must-have of Visibility |Looking beyond the single plant

The market today makes vast data and analytics capabilities available to nearly any industrial organization.

Nonetheless, most companies only periodically benchmark and diagnose performance within and across plants when they should be doing it on an ongoing basis.

Of course, not all KPIs are comparable across different plants making different things, but many are, especially downtime, quality, energy usage, environmental parameters, health and safety.

Optimizing the value chain, not to mention optimizing individual plants, cannot be done with simple quarterly or monthly “snapshots” alone.

Digital Twin technology

The third must-have of Visibility | To support decision making, look to the future with digital twins

The ability to look ahead, predict future performance, and perform what-if analyzes on various scenarios is essential to optimizing operations and the supply chain.

It is also the basis for planning and capital allocation to eliminate bottlenecks and add new products to existing plants (or, of course, new ones).

Manufacturing companies today can define digital twins of production process, supply chain and financial performance, and combine them together to support higher-level decision making, using artificial intelligence (AI) machine learning techniques.

Digital twins can identify and unravel relationships that are new, complex, and difficult to understand otherwise.

As former US Secretary of Defense Donald Rumsfeld has often said, they can reveal “the known unknowns and the unknown ones”.  

Recommendations

Every manufacturing company should – at a minimum – use available data on production, finance and supply chain to have better profitability immediately.

But if your company wants to take advantage of even the hidden opportunities, it must work to gain greater visibility into the data created within the production environment.

Critical early stages include:

  • REVIEW ENTERPRISE-LEVEL AND FACTORY-LEVEL IT AND OT ARCHITECTURES. This step is not a must-do action One-off: IT/OT architecture is an ever-evolving aspect of the organization.
  • CONSIDER A MULTI-LEVEL ANALYTICAL APPROACH. It is desirable to have one or more analysis tools that access different databases for different analyses. A variant of the operational architecture provides that the datalake is on top of all other databases to act as an aggregator.
  • DON'T NEGLECT THE OPTIONS AVAILABLE. There are many tools and solutions to choose from on the market. If a vendor can combine them into an integrated solution, so much the better. However, most businesses will use multiple tools.
  • DON'T LOSE SIGHT OF THE TARGET. Never forget your goals in terms of desired outcomes, and always keep in mind the use cases that drive your data access requirements and technology selection.

A technological project is not an end, but a means to achieve the desired results.

Learn how industrial companies create visibility in the webcast with LNS Research, “Driving Profitability with Plant and Process Data”.