Researchers Noga Chemo, Yaniv Mordecai, and Yoram Reich from the Department of Industrial Engineering at Tel Aviv University have introduced a groundbreaking framework designed to revolutionize the generation and analysis of safety requirements for complex safety-critical systems. Their work, titled “Foundational Analysis of Safety Engineering Requirements (SAFER),” integrates Model-Based Systems Engineering (MBSE) with Generative AI to address longstanding challenges in safety engineering.
SAFER is a model-driven methodology that aims to improve the generation and analysis of safety requirements, which are often specified by multiple stakeholders with uncoordinated objectives. This lack of coordination can lead to gaps, duplications, and contradictions that jeopardize system safety and compliance. Traditional approaches to this problem have been largely informal and insufficient for addressing these critical issues. The SAFER framework enhances MBSE by consuming requirement specification models and generating several key results. These include mapping requirements to system functions, identifying functions with insufficient requirement specifications, detecting duplicate requirements, and identifying contradictions within requirement sets.
One of the standout features of SAFER is its ability to provide structured analysis, reporting, and decision support for safety engineers. This structured approach ensures that safety requirements are thoroughly vetted and aligned with system functions, thereby enhancing the overall safety and reliability of the system. The researchers demonstrated the efficacy of SAFER using an autonomous drone system, where it significantly improved the detection of requirement inconsistencies. This demonstration highlighted the framework’s potential to enhance both the efficiency and reliability of the safety engineering process.
The researchers emphasize that Generative AI must be augmented by formal models and queried systematically to provide meaningful early-stage safety requirement specifications and robust safety architectures. By integrating Generative AI with formal models, SAFER offers a comprehensive solution that addresses the complexities and challenges of modern safety engineering. This innovative approach not only improves the quality of safety requirements but also ensures that they are systematically analyzed and validated, leading to safer and more reliable systems.
In practical applications for the marine sector, the SAFER framework could be a game-changer. Maritime systems, such as autonomous ships and offshore platforms, are highly complex and require stringent safety measures. The integration of SAFER could help identify and mitigate potential safety issues early in the design phase, ensuring that these systems meet the highest safety standards. By providing structured analysis and decision support, SAFER can enhance the safety engineering process, making maritime operations safer and more efficient. Read the original research paper here.