Getetec GmbH
Home Competencies Imprint Privacy
EN | DE

AI for Systems Engineering

Intelligent automation for model-based development projects

Artificial intelligence is changing the way we develop systems. But the real value comes not from AI as an end in itself, but from its targeted integration into existing engineering processes and tools. My focus is on deploying AI technologies in ways that tangibly improve day-to-day work in MBSE projects – from automated model creation to intelligent quality assurance.

Why AI in Systems Engineering?

Model-based systems development generates large volumes of structured data: requirements, function trees, logical architectures, interface specifications, traceability matrices. Many tasks in this domain are rule-based, repetitive, or require processing extensive contextual information – precisely the strengths of modern AI systems.

Typical application areas:

  • Requirements analysis – Automatic detection of ambiguities, missing acceptance criteria, or inconsistencies in requirement texts. AI models can classify requirements, identify duplicates, and generate suggestions for structuring according to RFLP levels.
  • Model generation – From textual descriptions, existing documents, or legacy data, AI can generate initial model structures: function hierarchies, interface specifications, or logical architectures as a starting point for refinement by the engineer.
  • Consistency checking – Automatic validation of models for traceability completeness, correct interface typing, and compliance with modeling guidelines – significantly faster and more reliable than manual reviews.
  • Knowledge extraction – Existing documentation, standards, and guidelines can be processed by AI and translated into context-aware modeling recommendations.

My Technology Approach

Integration, not isolated solutions

I develop AI automation solutions that integrate seamlessly into existing toolchains. I pursue two complementary directions:

AI for PREEvision – As a long-standing PREEvision expert, I know the data model, API interfaces, and typical workflows in automotive and non-automotive E/E development projects. I leverage this expertise to develop AI-powered automations that work directly with PREEvision models: automated model analyses, rule-based consistency checks, AI-assisted generation of model content from natural language descriptions.

AI for RFLP and open SE tools – In the context of my SE-Master workbench and the FlowSpec DSL, text-based modeling opens up unique opportunities for AI integration. Large Language Models (LLMs) can directly read, analyze, and generate FlowSpec models – without the detour through proprietary binary formats. This enables a natural, dialog-oriented interaction with system models: engineers describe changes in natural language, and the AI translates them into valid model structures.

Automation with n8n and modern workflow engines

Beyond direct AI integration, I use workflow automation with tools like n8n to orchestrate recurring engineering processes. Typical scenarios:

  • Automatic import and classification of new requirements from external sources
  • Regular model validation with report generation
  • Notification chains for model inconsistencies
  • Batch processing of documents to extract modeling-relevant information

GDPR-compliant processing

For projects with sensitive data, I offer solutions that can be operated entirely on-premises – on your own hardware, without cloud dependency. Local AI models ensure that no project data leaves the corporate network.

Service Portfolio

  • Potential analysis – Identification of the most impactful AI use cases in your SE process. Which tasks can be automated? Where is the greatest leverage for quality and speed?
  • Prototyping & proof of concept – Rapid implementation of initial AI automations as feasibility demonstrations. Within a few weeks, a working prototype emerges that demonstrates concrete value for your team.
  • Production deployment – Transition of successful prototypes into robust solutions integrated into daily workflows. Including monitoring, error handling, and documentation.
  • Technology consulting – Vendor-neutral evaluation of AI technologies for your context: which models, which infrastructure, which operating mode (cloud vs. on-premises) fits your requirements for data privacy, performance, and budget?