Software-defined / virtualised DCS & PLC

Control Systems — DCS · PLC · ESD · HMI Over-Promise Near-term · 2–5 yrs Evidence E3 · First production use

Control software that runs on standard servers or industrial PCs instead of proprietary controller hardware. The first open-architecture plant and server-based controllers from mainstream suppliers are now in production; the long-term support model is still unproven.

EMERGINGPROMISEREALITYACCEPTANCEPROVEN
Software-defined / virtualised DCS & PLC is placed at the Over-Promise stage, with an estimated time to proven use of 2–5 yrs (near-term) and evidence level E3, first production use. Stage reflects attention versus results; the evidence level reflects deployment proof. The height of the curve is illustrative only. October 2026 edition.

Separating control software from hardware

A conventional DCS controller or PLC is a closed product: processor, real-time operating system and control runtime are designed, certified and supported together. When the hardware becomes obsolete the owner usually replaces the controller and often migrates the application, even though the control strategy has not changed. Over a plant's 20- to 30-year life this can happen several times.

Software-defined control keeps the runtime but moves it onto general-purpose computing, as a virtual machine or a container on a real-time-capable host. Redundancy becomes a function of the hosting layer, controllers can be created from templates and backed up as images, and I/O is reached over the network. There are two distinct routes. In vendor-integrated platforms one supplier provides the runtime, the hosting stack and support. In open architectures, such as the Open Process Automation Standard (O-PAS), components from different suppliers are combined by the owner or a system integrator.

Two routes to software-defined control
Vendor-integratedOpen architecture (O-PAS)
Who integratesThe control-system supplierThe owner or a system integrator
Examples in productionSiemens S7-1500V at Audi; Emerson DeltaV IQExxonMobil Baton Rouge Lighthouse
Main benefitHardware independence with one point of supportHardware independence and component-level competition
Main open questionLock-in moves to the software layerWho carries lifecycle responsibility

What is actually running

ExxonMobil's Open Process Automation "Lighthouse" project at its resins finishing plant in Baton Rouge, Louisiana, is the most cited open-architecture example. It is built on O-PAS version 2.1, with distributed control nodes running IEC 61499 runtimes, about 1,000 I/O points, and a system integrator rather than a DCS supplier responsible for support. It passed factory acceptance testing in August 2024, was cut over during a scheduled outage in November 2024, and produced its first saleable product on 18 November 2024 12. Reports from early 2025 describe stable operation since start-up, about a week of operator training because the HMI graphics were kept unchanged, and a next project of around 200 I/O 1. Published accounts differ on whether the system has around 100 controllers or around 100 control loops 12; we have used the I/O count, on which they agree.

On the vendor-integrated side, Siemens describes its SIMATIC S7-1500V as the first TÜV-certified virtual fail-safe controller. It runs on Audi's Edge Cloud 4 Production infrastructure on the e-tron GT line at Böllinger Höfe, after roughly three years of requirements definition and testing 3. Emerson's DeltaV IQ controller packages real-time control and redundancy in virtual machines on hyperconverged servers. Emerson introduced it with DeltaV version 15 Feature Pack 3 in March 2025 4 and, at its user conference in May 2025, said that a pair of servers can run up to 36,000 I/O, against about 1,500 for a hardware controller 5.

The hard questions are about the life cycle

The technical questions are reasonably well bounded: real-time performance on the chosen hardware, deterministic network paths to I/O, and failover behaviour when a host fails. Suppliers and early adopters have answered them for the configurations they have tested.

The harder questions are about the life cycle. A hardware controller is supported as one unit for decades. A software-defined controller depends on a hypervisor, an operating system, server firmware and network switches, each with its own patch and end-of-life schedule, typically far shorter than the plant's. Somebody has to own compatibility testing of every combination. In an IT data centre that is routine; in an OT environment that must run for years between turnarounds it is a new discipline. Cybersecurity changes shape for the same reason: the hosting layer becomes part of the control system and must sit inside the same zone design, patching regime and monitoring.

There is also a skills shift. Commissioning a controller now includes server, virtualisation and network configuration, and a fault may sit in any layer. Owners considering the move should be honest about whether their own OT team, the supplier or an integrator will hold that knowledge for the next twenty years.

Where to start

The sensible early applications are those where flexibility has obvious value and the consequences of a mistake are contained: simulation and operator training systems, migration of obsolete controllers where the application can be preserved, modular and batch plants, and new units with non-critical control. Safety systems are a separate question. Certified virtual fail-safe control has appeared in discrete manufacturing; process-industry safety instrumented systems are likely to stay on dedicated hardware for some time.

Evidence timeline

  1. Aug 2024ExxonMobil OPA Lighthouse passes factory acceptance testing. 1
  2. Nov 2024Lighthouse cut over and producing saleable product from 18 November. 2
  3. Mar 2025Emerson releases DeltaV v15 Feature Pack 3, including the server-based IQ controller. 4
  4. May 2025Emerson states DeltaV IQ can run up to 36,000 I/O on a pair of servers. 5
  5. 2025Siemens describes the TÜV-certified S7-1500V virtual controller in use at Audi Böllinger Höfe. 3

Why it sits at the Over-Promise stage

Real systems are now in production, and supplier marketing has run well ahead of them. We keep the entry at the Over-Promise stage because the number of production plants is still very small, no software-defined system has yet been through a full hardware refresh in a process plant, and expectations of cost savings are ahead of the evidence. The 52 percent hardware and software saving often quoted for the ExxonMobil approach comes from an earlier analysis rather than measured project results 1, and should be read that way.

We expect the near-term horizon to hold for new projects and controller migrations. Replacement of the installed base will take much longer.

Adoption signals to watch

Questions to ask before committing

Terms used on this page

Virtual PLC
A controller runtime delivered as software to run on standard computing hardware.
IEC 61131-3
The standard that defines the common PLC programming languages.
IEC 61499
A standard for event-driven, distributed function-block applications, used by some software-defined platforms.

Sources

  1. New insights: 100-controller ExxonMobil Open Process Automation Control Engineering, Mar 2025
  2. Plug-and-play punches in Control Global, Mar 2025
  3. Factory automation: virtual PLC at Audi Siemens, n.d.
  4. Emerson enhances control system software to increase operational agility Emerson, Mar 2025
  5. Innovations sparkle at Emerson Exchange 2025 Control Global, May 2025

Related technologies

See this entry on the full maturity map · All Control Systems entries · How placements are made. Stage and time-horizon placements are editorial judgements based on the public sources listed above, as of October 2026. They are not engineering, safety or investment advice; see the Terms of Use.