AI-Driven Generation of SysML Diagrams from Requirements

Overview

In this Proof of Concept (PoC), we demonstrate the integration of the KIMBA AI Chatbot directly into professional modeling environments like MagicDraw. The goal is to bridge the gap between textual requirements and formal system models, significantly accelerating the early stages of automotive development.

Key Functionalities

The demonstration focuses on two critical SysML (Systems Modeling Language) capabilities:

• Activity Diagrams: Automatically generating behavioral flows from complex requirement sets (e.g., "Battery Charging" processes).
• Block Definition Diagrams (BDD): Defining system structures and architectural hierarchies through simple AI prompts.

Process & Integration

By simply pasting raw requirements - such as those provided by project partners like BMW - into the chatbot interface, the AI interprets the logic and renders a digital model in seconds. In this specific use case, the chatbot mapped out the entire "Battery Charging" sequence, from the customer approaching the charging port to the LED status signals.

Strategic Value

The automated generation of models serves as a Single Source of Truth, enabling:

• Early Error Detection: Finding logic flaws before a single line of code is written.
• Cost Efficiency: Reducing the need for expensive late-stage corrections or vehicle recalls.
• Automated Testing: Deriving software test cases and simulations directly from the AI-generated models.

KIMBA proves that AI doesn't just assist—it accelerates the engineering lifecycle while maintaining high precision in safety-critical automotive systems.