Combining Event Tree Analysis and System-Theoretic Process Analysis for Hydrogen Refueling Station Safety

Hydrogen refuelling stations are complex socio-technical systems where safety outcomes emerge from interactions among technical components, human operators, and organizational structures. Traditional risk assessment methods, such as the Event Tree Analysis model, analyze accident sequences and estimate probabilities of hydrogen releases; however, they struggle to capture nonlinear dependencies, timing effects, and feedback dynamics. This paper applies System-Theoretic Process Analysis together with Event Tree Analysis to the same hydrogen compressor leak scenario to identify gaps in traditional event-based risk assessment and explore how the two methods can be meaningfully combined. Event Tree Analysis maps accident progressions, and System-Theoretic Process Analysis identifies unsafe control actions, feedback delays, and organizational factors influencing barrier performance. Linking System-Theoretic Process Analysis causal factors to Event Tree Analysis sequences illustrates how technical, human, and organizational interactions influence accident progression. Combining these methods provides a better way to assess hydrogen safety risks and capture both direct and emergent contributors to accidents.

Keywords: Event Tree Analysis, System-Theoretic Process Analysis, Hydrogen Safety, Risk Assessment, SocioTechnical Systems, Systems Thinking.