Features
From digital shadow to a twin you can simulate.
Live data mirrored onto a model is a digital shadow. It becomes a twin when the model, the data and the process logic are linked closely enough to test changes on. SteamForge gives you each layer as an API resource.
- Twins
- Scenes
- Models
- Connectors
- Simulations
- Scenarios
- API keys
Line 3 · live
Synced 4s ago · 42 assets
- Throughput (shift)
- 287
- on plan
- OEE
- 78.2%
- -1.4 vs last week
- Bottleneck
- ST-07
- queue 14
- Assets online
- 41/42
- 1 offline
| Station | State | Cycle |
|---|---|---|
| ST-04 Press | Running | 42.0 s |
| ST-05 Robot cell | Running | 39.5 s |
| ST-07 Welding | Starved | 58.2 s |
| ST-08 Inspection | Offline | — |
Core capabilities
3D scene modeling
Turn layouts into a navigable scene that the rest of the platform hangs off.
- Import OpenUSD, glTF and STEP
- Tag assets, stations, conveyors and zones
- Scene versions you can diff and roll back
Equipment simulation
Attach behavior to machines, from simple cycle times to physics models.
- FMI 2.0 / 3.0 FMUs for co-simulation
- Reduced-order models for real-time runs
- Parameter sets per asset type
Process simulation
Discrete-event simulation of material flow across a line, plant or warehouse.
- Shift, week or custom horizons
- Breakdowns, changeovers and staffing rules
- Throughput, utilization and queue output
Live state visualization
Map telemetry onto the scene so operators see the plant as it runs.
- OPC UA, MQTT and Modbus TCP via edge connector
- Server-Sent Events stream for your dashboards
- Thresholds and webhooks on any signal
Virtual testing
Test control logic and process changes against the twin before the floor.
- Run scenarios against recorded or live state
- Repeatable seeds for regression testing
- Virtual commissioning workflows on Enterprise
What-if scenarios
Fork a twin, change one thing, and compare against the baseline.
- Layout, cycle time and buffer changes
- Side-by-side result comparison
- Batch runs for parameter sweeps
Where teams use it
Ordered by the asset lifecycle, from before the line exists to running it day to day.
- 01 Virtual commissioning Exercise control sequences and interfaces against the twin before equipment is installed.
- 02 Line and process simulation Validate takt time and material flow for a new or changed line.
- 03 Bottleneck and capacity analysis Rank stations by queue time and see where an extra shift or machine pays off.
- 04 Condition monitoring Watch live state against expected behavior from the equipment model.
- 05 Energy and logistics Model AGV routes, buffers and energy draw across a facility.
- 06 What-if planning Try layout moves and capacity expansions before committing capital.
Open standards in, open formats out
No proprietary scene format to learn. Bring models you already have, and pull results into your own tools.
- OpenUSD
- Scene description
- glTF / STEP
- Geometry import
- FMI 2.0 / 3.0
- Model exchange
- OPC UA
- Industrial data
- MQTT
- Telemetry
- Modbus TCP
- Legacy devices
- REST + SSE
- API and streaming
- Webhooks
- Events out
Data security and privacy
Operating twins touch OT systems like PLCs, MES and SCADA, so hosting, access and network separation are settled up front, not after the pilot.
Outbound-only edge connector
Runs inside your network and opens no inbound ports. Read-only by default.
Encryption
TLS 1.2+ in transit and AES-256 at rest for scenes, models and state history.
Your data stays yours
Scenes, models and results belong to you. We do not use customer data to train models.
Retention you control
Set state history retention per twin and delete a twin and its data with one call.
Scoped API keys
Keys per workspace and environment, with read-only keys for dashboards.
Dedicated deployment
Enterprise can run in a dedicated environment or on your own infrastructure.
Getting started
- Step 1
Get a Sandbox key
Email us and we send a free key for one twin.
- Step 2
Upload a scene
POST an OpenUSD or glTF layout to /v1/scenes.
- Step 3
Connect data
Install the edge connector or push state over REST.
- Step 4
Run a scenario
Start a simulation and read results from the API.