Workflow-Level Design Principles for Trustworthy GenAI in Automotive System Engineering

TL;DR

This paper presents workflow-level design principles for integrating trustworthy GenAI into automotive system engineering, emphasizing traceability, correctness, and safety in safety-critical applications.

Contribution

It introduces novel workflow principles and practical methods for safe, traceable GenAI use in automotive engineering, including delta propagation, model validation, and regression testing.

Findings

Section-wise decomposition improves specification completeness.

Requirement deltas are effectively propagated into SysML v2 models.

Traceable regression testing enhances safety in GenAI-assisted engineering.

Abstract

The adoption of large language models in safety-critical system engineering is constrained by trustworthiness, traceability, and alignment with established verification practices. We propose workflow-level design principles for trustworthy GenAI integration and demonstrate them in an end-to-end automotive pipeline, from requirement delta identification to SysML v2 architecture update and re-testing. First, we show that monolithic ("big-bang") prompting misses critical changes in large specifications, while section-wise decomposition with diversity sampling and lightweight NLP sanity checks improves completeness and correctness. Then, we propagate requirement deltas into SysML v2 models and validate updates via compilation and static analysis. Additionally, we ensure traceable regression testing by generating test cases through explicit mappings from specification variables to architectural ports and states, providing practical safeguards for GenAI used in safety-critical automotive engineering.