Machine Synthetic Data Generator¶
1. Product Description¶
1.1. Solution Overview¶
AI-MDG (Artificial Intelligence-Machine Synthetic Data Generator) is a cloud-hosted simulation platform developed by IANUS Simulation GmbH. It enables engineers and decision-makers to perform high-fidelity simulations on digital representations of physical systems—so-called Digital Twins—without requiring local installation or specialized hardware on the client side.
The platform combines state-of-the-art computational methods with a user-friendly interface to support the entire simulation lifecycle, from model upload to result interpretation. AI-MDG is designed to handle complex, multi-parameter simulation scenarios, with particular focus on fluid dynamics, extrusion processes, and thermal behavior in industrial applications.
Delivered via a web browser, AI-MDG empowers users with scalable compute resources, including GPU acceleration and HPC backend compatibility, while ensuring accessibility, maintainability, and data security through centralized deployment.
1.2 Features¶
🔹 Web-Based Access¶
No installation required
Cross-platform compatibility (Windows, macOS, Linux)
Accessible through all major modern browsers
🔹 Digital Twin Management¶
Upload and manage 3D models (STEP format)
Interactive 3D viewer with rotation, pan, and predefined views
Version control and simulation history tracking
🔹 Simulation Engine¶
High-performance backend supporting complex CFD and thermodynamic simulations
Customizable input parameters: material, temperature, throughput, etc.
Simulation status tracking with live progress indicators
🔹 Automated Report Generation¶
Downloadable PDF reports for each completed simulation
Includes detailed process data, mesh statistics, inflow/outflow behavior
Graphical plots: pressure, velocity, viscosity, shear rate, residence time, and more
🔹 Interactive Visualization¶
Real-time 3D rendering of geometry and flow conditions
Advanced data visualizations including streamlines and cut planes
1.3. Prerequisites¶
• Hardware Requirements¶
To ensure optimal performance of the AI-MDO system, a compute environment equivalent to the JURECA Pre-Exascale Supercomputer is required. The recommended hardware configuration includes:
CPU: 2× AMD EPYC 7742 (2× 64 cores, 2.25 GHz)
Memory: 512 GB DDR4 RAM (16× 32 GB, 3200 MHz)
GPU: 4× NVIDIA A100 GPUs (each with 40 GB HBM2e)
Networking: 2× InfiniBand HDR interfaces (NVIDIA Mellanox ConnectX-6)
Note: Lower-tier hardware may result in significantly reduced performance or unsupported execution of AI-MDG workloads.
• Software Requirements¶
No additional software installation is required on the client side. The AI-MDO platform is accessed entirely via a modern web browser. We recommend using the latest version of one of the following:
Google Chrome
Mozilla Firefox
Microsoft Edge
Note: JavaScript and WebGL support must be enabled in the browser.
• External Dependencies¶
A valid user account issued by IANUS Simulation GmbH
An active and stable connection to the public internet
2. Installation¶
No installation is required to use the AI-MDO system. The application is fully web-hosted and is delivered as a Software-as-a-Service (SaaS) platform.
Users can access the system exclusively through a supported web browser. There are no client-side components, plugins, or local deployments necessary.
Note: Ensure that your browser is up to date and meets the requirements listed in the Software Requirements section.
3. Initial Configuration¶
3.1. First Steps¶
• Login¶
Upon navigating to the AI-MDO web application, users are greeted with a login screen. After successful authentication, the user is redirected to the main dashboard, which provides an overview of the most recent simulations and digital twins.

3.2. Main Workflows¶
• Simulations Tab¶
Each simulation is represented by a geometry tile, with a status indicator in the top-right corner:
✅ Check mark: Simulation completed successfully.
❌ Cross mark: Simulation was cancelled or failed.
⏳ Hourglass: Simulation is currently in progress.
To start a new simulation:
Click “New Simulation” (top-right or sidebar)
Enter:
Simulation Name
Select Digital Twin
Inflow Temperature [°C]
Material
Throughput [kg/h]
Click Next to review input
Click Start Simulation
A success or error message will appear after submission. Progress can be monitored in the Simulations tab.
¶
• Digital Twins Tab¶
Users can manage digital twins through a 3D geometry viewer.
Interaction:¶
Left-click + drag: Rotate
Right-click + drag: Pan
Home button: Reset view
Cube icon: Top, bottom, front, back, left, right views
To exit the viewer, click the “X” icon.
Create a New Digital Twin:¶
Click “New Digital Twin”
Upload a STEP file via drag-and-drop or file browser
Name the digital twin
Click Create
A confirmation message appears upon success. The twin is now available for simulation.
¶
• Viewing Results¶
Click on a ✅ simulation to access its results:
Simulation metadata
3D model preview
Input parameters
Under the Results tab, download the PDF report.

Report Includes:¶
Process Data: Temperature, Throughput, Mesh Cell Count
Inflow: Material, Inlets, Melt Density, Pressure
Outflow: Velocity, Cell Count
Visuals:
Pressure & Velocity Plots
Mesh & Streamlines
Viscosity and Shear Rate
Residence Time & Distribution
Material Flow Paths

4. General Queries¶
4.1. Installation and Configuration Contact (If Service Provided)¶
For assistance with setup, credentials, or login issues:
Contact: support@ianus-simulation.de
4.2. Technical Support¶
Company: IANUS Simulation GmbH
Website: https://ianus-simulation.de/en/
Logo:
4.3 Licensing and Commercial Terms¶
AI-MDO is available under the following models:
AGPLv3 (open source)
Private/Commercial License
Subject |
Value |
|---|---|
Payment Model |
Available on quotation |
Price |
Available on quotation |
Contact IANUS for licensing and pricing details.
5. Appendices¶
5.1. Glossary of Terms¶
Term |
Definition |
|---|---|
AI-MDO |
Simulation platform for Digital Twin-based design optimization |
Digital Twin |
Virtual representation of a physical component or system |
Simulation |
Numerical model used to predict real-world behavior |
STEP File |
3D geometry exchange file format |
3D Viewer |
Interactive tool to inspect geometries |
Mesh |
Computational grid for simulation |