What is virtual commissioning?
9 min read · By neexo engineering team · Vejle, Denmark
Published: 2 July 2026
Virtual commissioning connects PLC logic to a simulated machine model so you can test sequences, safety interlocks, and operator flows before hardware is wired on the shop floor. Machine builders use it to find integration faults earlier, shorten FAT rework, and give customers confidence that the control system behaves as designed.
Why do machine builders use virtual commissioning?
Commissioning risk shows up late. A sequence that looked fine in the PLC editor fails when sensors, actuators, and operator screens are wired together. Travel costs stack up when customer representatives fly in for a FAT that turns into a debug week. Virtual commissioning moves a large share of that integration work earlier, while changes are still cheap.
Teams adopt it for three recurring pressures: shorter delivery windows, more variant complexity, and customers who want evidence before they sign off on hardware. When logic is tested against a representative model, project meetings shift from guessing to demonstrating behaviour. That changes how you plan FAT, service documentation, and operator training.
The pattern repeats across industries: packaging lines with frequent recipe changes, custom assembly cells with tight safety zones, and machines where mechanical design iterates in parallel with software. In each case, the cost of a late logic fault is measured in overtime, airfares, and customer trust: not just engineering hours.
Sales teams can use recorded virtual runs in technical review meetings without promising acceptance outcomes. That separation keeps commercial conversations honest while still showing progress.
It is not a shortcut around discipline. You still need clear test cases, version control, and ownership of who updates the model when mechanics change. The payoff is fewer surprises when the real machine powers up, and a FAT agenda that reads like verification instead of discovery.
- Catch sequence and interlock errors before panels are fully wired
- Run customer review sessions without blocking the build hall
- Give service and training teams a stable environment before install
How does virtual commissioning work in practice?
A typical setup links three layers: the control program, a machine model, and a test plan. The PLC code, or a soft-PLC runtime, executes as it would on the target controller. The model responds to outputs and feeds inputs back: limits, delays, fault states, and motion where relevant. Test cases describe what “good” looks like for each mode, alarm, and recovery path.
Engineers start with scope. Not every axis needs photoreal graphics on day one. Many projects begin with critical sequences: safety circuits, mode changes, recipe handling, and HMI navigation. As confidence grows, teams extend coverage to edge cases that are painful to reproduce on hardware, such as double-sheet detection, jam recovery, or starvation at upstream equipment.
Signal mapping is the quiet work that makes or breaks the setup. Each physical I/O address needs a twin equivalent: digital inputs from simulated sensors, analog values within realistic ranges, and handshake bits for upstream/downstream equipment. Document the map in the same revision system as the PLC project so drift is visible.
Sessions look like structured FAT rehearsals. One engineer drives the PLC, another operates the HMI, and a third logs deviations. Issues become tickets with reproduction steps: the same habit you want on the physical line. When the model and logic are aligned, you can rerun the full suite after each software drop in minutes instead of hours.
Integration with your existing toolchain matters. Export I/O lists from the PLC project, align tag names with the twin, and tag test cases to functional requirements or user stories if your project uses them. That traceability helps when a customer asks why a specific interlock was verified and which build proved it.
What is the difference between a digital twin and virtual commissioning?
The terms overlap in marketing slides but serve different jobs in delivery. A digital twin is a living representation of the machine, geometry, kinematics, signals, and often documentation links. It can exist for sales, training, or service without being tied to a commissioning milestone.
Virtual commissioning is the activity: proving that control software and the twin (or a lighter simulation) behave correctly together against defined tests. You can commission with a coarse model if the goal is logic validation. You need a richer twin if operators must practice realistic procedures or if sales configurator data must match shop-floor behaviour.
A twin without tests is a visual asset. Commissioning without a maintained model is a one-off exercise that ages badly. The durable approach connects them: the twin supplies physics and context; the commissioning layer supplies pass/fail criteria, regression runs, and traceability to contract specs.
In practice, machine builders benefit when the same asset backs multiple uses. A twin built for virtual commissioning can later support interactive manuals or service training, provided someone owns updates when engineering changes the design.
Virtual commissioning does not replace physical FAT; it reduces surprises at FAT.
Which PLC platforms support virtual commissioning?
Most major automation stacks support some form of virtual testing. Siemens offers PLCSIM Advanced and integrated workflows with TIA Portal. Rockwell users work with Emulate and Studio 5000. Beckhoff, B&R, and CODESYS-based systems expose soft-PLC or hardware-in-the-loop options depending on runtime and I/O coupling.
You will choose between PLC-centric simulation and coupled 3D plus PLC. PLC-only tools are fast for logic, timers, and alarm paths; they struggle when spatial behaviour, operator sight lines, or mechanical clearance matter. Full 3D plus PLC costs more to set up but surfaces clashes that pure logic simulation misses, for example, an arm trajectory that trips a light curtain in the model but not in a table-driven stub.
Hardware-in-the-loop (HIL) sits between the two: real controllers with simulated I/O for high-fidelity timing tests. CAD-heavy teams often import mechanical data into Unity or similar runtimes for motion and operator context. Fieldbus behaviour and scan-cycle effects sometimes only appear at this level.
At neexo we typically pair a Unity-based 3D flow with live or soft-PLC code so sequences, HMI screens, and spatial constraints are exercised together, not as separate checklists. Platform choice should follow your FAT risks, not the reverse.
- PLC-only: best for early logic, interlocks, and alarm handling
- 3D plus PLC: best when motion, access, or operator procedure matters
- HIL: useful when cycle time and fieldbus timing must match production controllers
When is virtual commissioning worth the investment?
The business case is strongest when FAT risk is high relative to model cost. Custom machines with many variants, tight acceptance dates, or distributed teams usually see payoff quickly. If a single FAT overrun costs more than building a focused twin, the math is straightforward.
Compare scenarios with your own rates: customer daily standby, rework labour, and opportunity cost of a delayed ship date. Even modest reductions in FAT rework days can justify simulation when multiple stakeholders travel to witness acceptance.
It is harder to justify for one-off simple cells with stable logic and local acceptance; though even then, a lightweight simulation can pay back if travel or rework is likely. Start with the top ten failure modes from your last three projects. If those faults could have been triggered in software, you have a scope.
Budget for maintenance. A twin that drifts from the as-built machine erodes trust. Assign an owner to update the model when BOMs, sensors, or HMI flows change. Teams that treat the twin as a project deliverable, not a demo, keep it useful through FAT and into service.
On multi-machine programs, standardize virtual FAT templates per product family. Shared case libraries cut setup time on variant three and four, where customers expect the same rigour as the first unit.
What results should you expect?
Expect fewer logic defects at physical startup, not zero work on the floor. Virtual commissioning removes whole classes of bugs: wrong interlock order, missing alarm acknowledgements, HMI states that do not match PLC modes, and recipe parameters that never got wired through. Mechanical and electrical issues still appear where the model was simplified.
Schedule impact varies. Early projects invest time building the model; later projects reuse templates, test libraries, and signal maps. Customer-facing teams often report smoother FAT openings because acceptance criteria were demonstrated weeks earlier. Service teams gain a training surface before site install.
Quality records improve when test runs are logged against software versions. Instead of debating what was tested, you export a run list: build number, cases executed, failures, fixes, reruns. That audit trail supports ISO-style project documentation and post-mortems when something still slips through.
Measure what matters to your stakeholders: defects found before ship, FAT days spent on rework, and repeat trips. Pair qualitative feedback from operators with counts from your issue tracker. That evidence feeds the next bid and helps you decide how much simulation to scope on the next machine.
Set expectations with sales and project management up front: virtual commissioning improves odds; it does not certify shipment dates by itself. When everyone shares that framing, results read as earned progress instead of missed promises.
- Fewer critical logic surprises during physical commissioning
- Clearer FAT agendas with pre-run pass/fail evidence
- Reusable test assets for variants and follow-on lines
See how neexo runs virtual commissioning
We build Unity-based twins linked to your PLC and HMI so your team can rehearse FAT scenarios before hardware is fully assembled.
Explore the serviceFrequently asked questions
Does virtual commissioning replace on-site commissioning?
No. Virtual commissioning reduces logic and integration risk before hardware is complete; physical commissioning still validates wiring, sensors, pneumatics, and real-world timing. Think of it as shifting debug work earlier, not eliminating the final install. Most teams run virtual sessions through FAT preparation, then use the same test cases as a checklist on the machine. The goal is fewer surprises and shorter critical-path time on the floor, not skipping safety or mechanical verification.
How accurate does the 3D model need to be?
Match fidelity to the test goal. For interlocks and mode logic, simplified geometry and representative signal timing are often enough. For operator training, clearance checks, or vision-related behaviour, invest in accurate kinematics and layout. Over-building graphics before core sequences pass wastes budget; under-building hides spatial faults. Review scope with whoever owns FAT acceptance. Update the model when engineering releases change mechanics or sensor placement so results stay credible for customers and internal sign-off.
Can we connect our real PLC program to the simulation?
Yes, on most platforms. Approaches range from soft-PLC runtimes in the engineering environment to hardware-in-the-loop with physical controllers and simulated I/O. The right choice depends on cycle-time sensitivity, fieldbus details, and whether remote stakeholders need to watch sessions. Early in a project, soft-PLC coupling is common for speed. Closer to FAT, teams often tighten fidelity. Document which runtime was used for each test run so results are comparable across software versions.
Who should own the digital twin after FAT?
Assign ownership before the project peaks, not after handover. Engineering usually owns signal maps and logic alignment; service or aftermarket teams benefit when they can train on the same asset. Without an owner, twins rot when CAD revisions land. A practical split: project engineering maintains accuracy through FAT, then transfers update rules to service with a named contact. Include twin access in handover packages alongside PLC backups and HMI exports so downstream teams can keep using it.
How long does it take to set up virtual commissioning?
A focused first scope (core sequences, safety paths, and primary HMI flows) often takes a few weeks for a custom machine, assuming PLC code and I/O lists are stable enough to map. Reuse accelerates follow-on projects: signal templates, test libraries, and standard motion blocks shrink setup time. Delays usually come from moving targets: late mechanical changes, unfinished HMI, or unclear acceptance criteria. Lock a minimum test list early and expand in iterations instead of waiting for a perfect model.
Related reading
A practical first step is to list the failures that burned you on the last FAT and mark which ones could have been caught with PLC-linked simulation. Bring that list to a short scoping call; we will help you decide model depth, platform coupling, and a test plan you can reuse on the physical machine.
Book a clarification call