Integrating Safety: Bridging STPA and MBSA with an Ontology-Driven Approach

On Friday, 28th November 2025, our external colleague Max Chopart gave talk about safety integration by bridging STPA and MBSA using ontology driven approach. Video including discussion and presentation slides are included in this report.

In this talk, Max presented a joint project between CTU and Airbus Protect that bridges the qualitative insights of System-Theoretic Process Analysis (STPA) with the Model-Based Safety Analysis (MBSA). He addressed the challenge of linking STPA’s natural language “Loss Scenarios” to MBSA’s formal “Observers” by developing an ontology-driven approach that maps scenarios directly to the MBSA model. Using a Cessna C510 battery system as a proof of concept, Max showed how the workflow can translate textual scenarios into executable boolean formulae for probability calculation. Finally, he outlined our future plans to enhance this process using Natural Language Processing to pre-annotate these scenarios for safety analysts.

Summary

Safety assessments in modern, complex systems require comprehensive approaches that account for both non-failure and failure scenarios. This presentation introduces an ontology-driven framework designed to efficiently integrate two powerful, yet distinct, safety analysis methodologies: System-Theoretic Process Analysis (STPA) and Model-Based Safety Analysis (MBSA). STPA offers a broad, hazard-based approach, capturing scenarios driven by systemic, human, and organizational factors. MBSA, conversely, excels at detailed failure-propagation analysis. While complementary, bridging these two requires a systematic mechanism to map relevant information.

Our core contribution is a domain ontology that acts as an intelligent interface with two key functions:

1. Filtering: It systematically analyzes STPA scenarios to identify and isolate those suitable for MBSA’s failure-based analysis.
2. Translation: It automatically converts these filtered STPA scenarios into MBSA-compatible “feared events” (observers), minimizing manual effort and error.

This systematic mapping allows us to leverage STPA’s wide-ranging hazard identification to seed a more focused and thorough fault-propagation analysis in MBSA.

This integration method streamlines the safety assessment process, enhances accuracy by reducing human intervention, and creates a more comprehensive, unified safety and reliability framework. Our early results demonstrate a more efficient and structured utilization of both STPA and MBSA’s strengths, critical for managing the complexities of contemporary safety-critical systems.

The presentation slides are available at this link.

Further Reading:

1. M. Chopart, J. Vidalie, A. Lališ, and K. Grötschelová, ‘Harmonizing Safety Perspectives: Integrating STPA outputs into MBSA for Comprehensive Aircraft Safety Assessment’, in 2024 New Trends in Civil Aviation (NTCA), Apr. 2024, pp. 261–267. doi: 10.23919/NTCA60572.2024.10517833.
2. M. Chopart, J. Vidalie, and X. D. Bossoreille, ‘Ontology-driven Integration of System-Theoretic Process Analysis and Model-Based Safety Analysis for Comprehensive Safety Assessment’, in 35th European Safety and Reliability Conference (ESREL 2025) and the 33rd Society for Risk Analysis Europe Conference (SRA-E 2025), Singapore EXPO, Singapore: Research Publishing Services, 2025, pp. 25–32. doi: 10.3850/978-981-94-3281-3_ESREL-SRA-E2025-P1191-cd.
3. A. Carniel, J. D. M. Bezerra, and C. M. Hirata, ‘An Ontology-Based Approach to Aid STPA Analysis’, IEEE Access, vol. 11, pp. 12677–12697, 2023, doi: 10.1109/ACCESS.2023.3242642.
4. M. Batteux, T. Prosvirnova, and A. Rauzy, ‘AltaRica 3.0 in 10 Modeling Patterns’, International Journal of Critical Computer-Based System, vol. 9, pp. 133–165, 2019, doi: 10.1504/IJCCBS.2019.098809.
5. M. Batteux, T. Prosvirnova, and A. Rauzy, ‘AltaRica 3.0 Language Specification’.