The role of Generative AI in simulation tool automation: From zero to hero

Addressing interface complexity

Crafting any CAE tool presents significant challenges in terms of user experience. As the software evolves with increasingly robust capabilities, its interfaces naturally become more complex, with expanding toolbars and deeper menu structures. This complexity creates two distinct challenges:

1. Novice users face a steep learning curve, when attempting to navigate the extensive feature set
2. Experienced users spend excessive time on repetitive tasks that require numerous clicks and manual operations

Automation journey

The evolution of simulation tools automation capabilities represents a journey from manual operations to intelligent assistance. This progression can be visualized as four distinct phases:

1. Manual clicking – Users performed all operations through direct interface interaction, requiring extensive knowledge of menu structures and feature locations
2. Code-based automation – Introduction of Python-based and Java-based scripting allowing recording and playback of operations
3. Embedded automation – GUI-based automation tools that reduced the need for programming expertise
4. GenerativeAI – The latest advancement featuring natural language processing that allows users to express simulation intent conversationally

This evolutionary path reflects Siemens’ commitment to continuously improving user experience while preserving the powerful capabilities of simulation tools.

Macro revolution

Automation through macros emerged as the first major solution to these challenges. In the simulation world, where users frequently perform iterative processes and complex setup sequences, macros provide several critical benefits:

• Efficiency: Reducing hours of manual setup to seconds of execution time
• Consistency: Eliminating human error in repetitive processes
• Reproducibility: Ensuring simulation setups can be precisely replicated

Simcenter products have consistently incorporated robust automation capabilities, often built upon scripting interfaces like Java or Python, enabling users to record and playback sequences of operations or develop custom scripts to streamline workflows.

Beyond traditional macros

While powerful, traditional Java macros present their own challenges:

• Programming expertise: Requires knowledge of Java programming
• Maintenance burden: Scripts need updats with software version changes
• Development time: Creating robust macros required significant time investment

The introduction of embedded automation represented the next evolutionary step. This approach allowed users to create automation sequences directly through the GUI, eliminating the need for programming knowledge. This approach significantly democratized automation capabilities, making them accessible to simulation engineers without programming backgrounds.

From structured interfaces to natural language interaction

Traditionally, simulation tools have required users to master their complexities through structured menus and toolbars that organize various functionalities. While this approach has significantly improved access to powerful features, it still requires users navigating, memorizing, and locating specific functions among numerous nested options. However, user interaction is evolving, and it’s intriguing to consider what might come after these rigid interfaces. Imagine being able to express your engineering challenges or simulation queries in natural language, with the software dynamically responding to them by carrying-out tasks and providing solutions. This development democratizes the use of advanced tools, making them accessible to users with varying levels of expertise.

The introduction of Generative AI into the realm of simulation software is a game-changer. By leveraging cutting-edge AI technologies, these tools can become more responsive to the intuitive queries and intentions of their users, not only as a smart consulter, but also as a smart assistant. Whether you’re a seasoned engineer or a novice experimenting with your first simulation, the GenAI-driven interface promises to make the process more seamless and less time-consuming.

Special thanks to Viorica Puscas, Roberta Luca, Gabor Schulz, Alexandru Grigore, Cosmin Ivan and Evelina Popa for bringing their technical expertise to this project.

Disclaimer

This is a research exploration by the Simcenter Technology Innovation team. Our mission: to explore new technologies, to seek out new applications for simulation, and boldly demonstrate the art of the possible where no one has gone before. Therefore, this blog represents only potential product innovations and does not constitute a commitment for delivery. Questions? Contact us at Simcenter_ti.sisw@siemens.com.