Going digital: 3 controls priorities

In this installment of Automation Zone, learn how your traditional maintenance strategy has to evolve as your plant’s technology does.

By Nathaniel Martin and Steve Copelin, GE Oil & Gas

Industrial organizations face a challenging paradox. Digital technology is advancing rapidly, while decades-old equipment is nearing obsolescence. To strike the right balance between adopting new technology and maintaining legacy equipment, the traditional maintenance strategy must evolve. As machines are updated and replaced with innovative new equipment and outfitted with sensors, maintenance is moving away from a focus on traditional hardware parts toward the digital assessment of machine health, software systems, and networks. This strategic shift will be critical for continued asset reliability, as research and advisory firm Gartner predicts that 26 billion devices will be "connected" by 2020. Machines must not only be running but also must remain online and communicate with other critical machines and operating systems across the plant. Current GE data suggests that 5% to 20% of assets have connectivity issues at any given time and that 2% to 3% percent of issues last longer than a month.

To ensure that maintenance strategy incorporates both machine health and continuous connectivity, operators should prioritize cybersecurity, system diagnostics, and communication.

1.    Cybersecurity: In today’s connected plant environment, cyber attacks are a concerning reality. Yet unlike in IT environments, operational technology (OT) environments can’t shut down and reboot to perform required updates and system patches. Machines that run 24/7 require specialized security testing and attention. Patches issued for industrial control systems (ICS) often have to be modified for live industrial environment to avoid complications, downtime, or vulnerabilities. Untested patches are unreliable without testing, because operators can’t predict how a particular system will respond to the updates. Validated patch management is a best practice that should be part of a regular maintenance strategy. With validated patch management, the patch is run in a virtual environment on-site or in a lab that mimics the plant environment to identify any incompatibilities before the patch is applied. This allows operators to determine what alterations need to be made to ensure uptime and protection against cybersecurity threats. Further, operators need to consider that multiple operating systems, including those for switches, central processing units (CPUs), and input/output (I/O) modules, exist within the OT environment. These systems must be patched as needed, and any anti-virus program running on control system devices should be monitored for necessary updates.

2.    System Diagnostics: Software is more advanced than ever, but it’s only as effective as the operators who run it and manage system diagnostics. Operators must understand the capabilities of each software package applied to machines and their control systems, as well as how to manage the data they provide. In the era of the industrial internet, it’s important to set a benchmark for healthy activity, recognize critical alerts and any anomalies identified via system analytics to mitigate potential risks, and identify trends to improve system performance. Engineers today know that if a data point is too far off the mean, something is happening, but now machines do the calculations. If it takes three seconds longer for a turbine unit to reach its standard temperature during startup than it did during the previous restart, which was also a few seconds behind the standard metric, an alert will go off and an operator can quickly recognize a shift in time stamp as an early indicator for failure. With more-exact insights into machine health, unplanned downtime can be avoided entirely, but plant engineers must be fluent in the new languages of system diagnostics to realize this value.

3.    Communication: Communication must be constant in connected plant environments. As the window to the industrial facility, human machine interface (HMI) upgrades are too often neglected, potentially endangering human safety, equipment, and operational output. Without critical upgrades, a network will not operate at optimum speeds. A slower network speed can prevent the system from obtaining real-time updates, resulting in delayed recognition and reaction to operational issues or complete equipment failures. Additionally, without updated equipment, operators may not know whether all backups are in place and working if a failure arises. Beyond paying heed to network speeds, operators must understand the flow of data across machines, systems and networks. A break in communication could indicate machine failure, a cyber attack or both. Fortunately, advanced software is intuitive enough to flag a communication breakdown as an error, and modern machines are built with redundancies so that if one switch fails, for example, another maintains communication and operations. System diagnostics then help operators quickly get to the bottom of the issue. Connectivity and communication are the foundation of digital industry, and as such, both must be maintained.

The Industrial Internet is projected to add $10 trillion to $15 trillion to the global gross domestic product in the next 20 years. This means industrial operators will rely more on digital systems to increase machine performance and reliability. Software, at the heart of these systems, provides previously unavailable insights into operations. It lets operators increase efficiency and reliability by helping generate data for performance analytics, analyzing systems before and during startup to avoid issues, and helping determine new alarm rationalization to save operators time. Shifting maintenance priorities to maintain and update industrial machine and control software, security, and connectivity allows engineers and operators to take advantage of technology advances.