One of the remaining obstacles to adopting O-PAS is that it likely requires users to learn some new and possibly unfamiliar IT-based skills, such as applying software containers, using Linux and related programming, and implementing virtual machine (VM) controls and others. However, acquiring these new capabilities is a small price to pay for the comparatively huge gains that O-PAS provides
For instance, because ExxonMobil completely removed its old DCS from the Baton Rouge plant, Dave DeBari, OPA technical team leader at ExxonMobil reported its support staff has been mastering some new skills, including:
- Cybersecurity in compliance with the IEC/ISA 62443 standard, which includes hardening and secure device onboarding; certificate authorization added to well-established, role-based access controls (RBAC); and deploying a zones-and-conduits network architecture.
Get your subscription to Control's tri-weekly newsletter.
- Automating the automation system with orchestration tools for automating lifecycle tasks; using systemness thinking to introduce new technologies; and employing automated workflows to simplify deployment when bringing systems online. These include support controllers, networks and software, and performing initial configurations, lifecycle maintenance and decommissioning.
- OPC UA for interoperability by communicating between control logic, I/O, HMI and historian functions using a standard information model.
- Advanced computing platform (ACP) as the OT data center, which lets users pick control devices and software with the size and performance level that’s most appropriate for the instantiations. The ACP also enables use of hyperconverged infrastructures, virtual machines (VM), containerized software and Kubernetes, and virtualized, software-based controllers with high availability.
- IT capabilities implemented with OT sensitivities, including Realtime Linux, a meshed network, firewall rules and virtual local area networks (VLAN).
“The network uses OPC UA to communicate with everything in the system, and lets us build the most suitable ACP for its process requirements,” adds DeBari. “We’re also using containerized software, Linux and VM controllers ready for time-sensitive networking. They’re working well, and managing machines, energies, chemistries and process operations. With ExxonMobil’s OPA Lighthouse running successfully, the O-PAS-based automation architecture has gone from being a crazy idea to what’s normal.”
Digitalize and virtualize
Even though O-PAS is already expansive and covers many areas, it can still overlap and fit into the larger worlds of digitalization, the Industrial Internet of Things (IIoT) and data analytics.
“We often don’t get the value from 3D scans and models because we don’t know how to store and centralize visual information well enough to make its useful content accessible to operators. If we did, we could develop new ways to work, and scale available technologies to allow more innovation and produce greater value,” says Michael Hotaling, innovation and digital strategy manager at ExxonMobil. “This is similar to the open-process automation (OPA)vision that OPAF had when it was first formed. We’re all seeking an ecosystem that relays data from open, physical assets to virtual, digital twins. However, all of this requires standardized interoperability that’s free of custom or proprietary automation, so more vendors can participate, and users can integrate better products with less effort. OPA also preserves each asset owner’s applications and data, and make sustainability cost-effective.”
Just as drivers expanded their perspective from rearview mirrors to global positioning systems (GPS), Hotaling reports that industrial work processes are moving beyond static models. This begins with taking data from reality, applying integrated engineering, using geospatial awareness to deliver information when and where it’s needed, ensuring interoperability, and accelerating the whole process with artificial intelligence (AI) and machine learning (ML).
“Users must ask if they’re doing digitalization for its own sake, or are they using it to improve how they’re working in the field,” explains Hotaling. “The process industry has had a ‘not built by us’ disease, and needs realize it doesn’t have to reinvent the wheel, so it can give workers what they need, where and when they need it. This transformation can begin by forming organizations like our Open-Asset, Digital-Twin working group. This can help asset owners and suppliers collaborate to separate hardware from software, and separate software from data in conjunction with OPA and digital-twin projects. These groups can also help members deploy agentic AI models in their ecosystems.”
