Simulate first. Commission second.
A digital twin is worth building when it changes a decision. We use simulation to find the bottleneck, size the system and de-risk start-up before the steel is ordered.
Plant data projects fail in a predictable way: the dashboards arrive before the process model, and nobody can say what the numbers mean. We work in the opposite order. Integrate the process first, build a model that reproduces the plant's real behaviour, and only then optimize against it.
That sequence pays off most at two moments. Before execution, when a simulation still costs less than a change order. And at commissioning, when the difference between a smooth ramp-up and a month of firefighting is whether the logistics were ever tested.
Inside the scope
Digital twins
Plant-level models built from the process design, used to validate hydraulics, energy and material flow before anything is installed.
Plant simulation
Throughput and capacity studies: scrap timing, ladle routing, heat scheduling and where the line will actually queue.
Automation
Control architecture and process models that carry the simulation work into the running plant rather than leaving it in a report.
Where we have done it
Insertec (2023): complete digital twin and simulation of an aluminium recycling plant, which validated the design before commissioning. Třinecké Železárny (Czechia, 2025): plant logistics simulation for a long-products maker, covering flow modelling, bottleneck detection and throughput optimization. Hyundai Steel (USA, 2024): FEED and basic engineering with full plant simulation and capacity validation ahead of the investment decision. Vinton Steel (USA, 2025): industrial digitalization study.
Is this the piece you are missing?
Send us the plant, the constraint and the timeline. You will get an engineer's answer, not a brochure.