The Benefits of Connecting Control Systems to Enterprise Systems

Strategically approaching system integration to improve operations—from linking controls to the enterprise to the Internet of Things and Big Data applications.

Andy Robinson, Avid Solutions Inc.
Andy Robinson, Avid Solutions Inc.

Shop floor to top floor system connections are slowly becoming more common. The idea of connecting control systems to enterprise systems has been around for a while, but mostly limited to specific projects or applications.

The most common example of how this is being done today is when a company creates a connection to an ERP system. Another common integration target for controls, especially in the chemicals and life sciences industries, is to Laboratory Information Systems. These systems will usually be relational databases, many times a SQL Server, which makes the integration much simpler.

An interesting side effect many manufacturers find after they create these system connections is that use cases blossom when other parts of the organization want to take advantage of the connectivity to perform a different function. Examples of this might be to exchange maintenance data with the shop floor or integrate operator training records into equipment security.

A combination of software and technologies are continuing to make the integration story easier and more cost effective. Integration to older ERP systems tended to be via formatted text files pushed around via FTP. Now, there are a variety of affordable and easy-to-use solutions for making direct calls to these ERP systems. Generic interface mechanisms like web services and more modern RESTful APIs have made enterprise systems much more accessible. Having these technically simpler interfaces, along with more capable SCADA platforms that can interact in a more native fashion, makes the integration portion a smaller part of the overall project’s cost and schedule.

When it comes to system integration for Internet of Things (IoT) applications, some manufacturers are adding limited technology, but few have adopted much of the available technology. Users are tending to pick and choose technologies that are being developed for IoT and using them in slightly different ways.

For example, we recently had a customer with a remote warehouse facility that needed to be monitored from the main campus. They were running an extension of their existing SCADA system over a wireless point-to-point radio, which wasn’t reliable. We explored the concept of using MQTT with a cloud-based broker to publish encrypted data from the warehouse and subscribe to the data in the SCADA system at the main facility. That is just one example of how IoT technologies might be repurposed for something other than their original use case.

These types of connections will likely become even more commonplace when they become more familiar to the traditional automation practitioner. A perfect example of this is with both Kepware and Wonderware writing MQTT drivers that live side-by-side with their traditional PLC drivers. Delivering data via this familiar technique allows engineers to focus on the value of the data and not the mechanics of data acquisition.

Then, of course, there is Big Data—another reason driving industrial systems integration. In my experience, only the most sophisticated customers are in a place where they can truly take advantage of Big Data. A common misunderstanding is that Big Data is going to take a pile of random data and put big red glowing circles around the really important things and then explain exactly what steps need to be taken to improve costs or reliability. The truth is, these software tools are there to help progressively sort through the correlations. It still takes deep knowledge of how a process functions before the results of the Big Data analysis can be turned into actionable information.

What is encouraging, however, is the rise of a number of specialty analytics software packages focused on the manufacturing industries. These packages tend to hide most of the complexity involved, leaving the engineer to focus on the most important function—the conceptual mapping of the models’ results to real world manifestations in their processes.

No matter the technique for improving your operations, the best way for facilities to start is to sit down and make the business case first. When making the business case, consider end goals such as cost reduction, improved traceability to support quality investigations or recalls, new product agility, reduced errors, etc. Once this justification is solidified, then you can work on identifying the gaps in your current processes and how these different technologies can be leveraged to close those gaps.

The best results are typically achieved when customers spend the time and money to develop a coherent, strategic plan and then attack each area in a very tactical way. If a manufacturer doesn’t clearly understand the potential benefits and perceives the cost of the project to be high, they are going to be fearful about even starting an exploration.

For more information on this subject, please view this webcast—“Linking Controls to the Enterprise in the Internet of Things Era.”

Andy Robinson is an Information Solutions Consultant at Avid Solutions Inc., a certified member of the Control System Integrators Association (CSIA). For more information about Avid Solutions, visit their profile on the Industrial Automation Exchange.

More in Data