Towards Productive Cyber Resilience and Safety Analysis in Model-Based Systems Engineering: a Quantitative Framework and Prototype Tool

Abstract

The integration of cyber resilience and safety analysis within Model-Based Systems Engineering (MBSE) remains a persistent challenge in defence and safety-critical systems engineering. Fragmented methodologies, tool silos, and the absence of standardised quantitative metrics impede productive, scalable analysis. This paper presents a prototype analytical framework and associated algorithms that operationalise practitioner-validated requirements for integrated cyber-safety analysis within MBSE. Drawing on a systematic literature review and semi-structured interviews with 18 domain experts, we derive five quantitative analysis strategies: Single-Pointof-Failure (SPOF) detection, channel security posture scoring, architectural dissimilarity assessment, automated Failure Mode and Effects Analysis (FMEA), and System Functional Hazard Assessment (SFHA). These strategies are combined into normalised composite indices- $S_{\text {cyber }}$ and $S_{\text {safety }}$-enabling earlystage comparative architectural screening. A worked example applied to a notional avionics control system demonstrates the utility of the approach. Preliminary findings provide promising preliminary indicators for design-space exploration and suggest that productivity improvements-defined as the reduction of manual cross-disciplinary translation and early error-detection cycles-are a plausible hypothesis. However, the measurement of actual productivity gains is outside the scope of this initial workshop paper and is explicitly planned for future industrial evaluation workshops.