Comparison of System Theoretic Process Analysis and Cause Tree Analysis Applied on an Autonomous Parking System from Safety of the Intended Functionality Perspective
Okan Özçetin, Pascal Brudke · 2024
The advancement of technology has led to a notable enhancement in the capabilities of intelligent vehicles and automated driving. This development has resulted in increased complexity, which in turn has given rise to a number of new risks. These risks are caused by automated driving functions of functional insufficiencies, performance limitations and driver misuse. Safety of the intended functionality (SOTIF) addresses the absence of unreasonable risk due to functional insufficiencies and performance limitations in accordance with ISO 21448:2022. As complexity increases, it becomes more challenging to identify functional insufficiencies and triggering conditions. To address this, various analytical techniques have been proposed in ISO 21448:2022, including inductive, deductive, and exploratory analysis. This paper concentrates on a comparison of systematic theoretic process analysis (STPA) and cause tree analysis (CTA) methodologies for automated parking system (APS) to determine triggering conditions. STPA methodology for SOTIF incorporates 4 steps. These are defining purpose of the analysis, modelling of the structure, unsafe control actions and causal scenarios (triggering conditions). For other methodology, cause tree analysis is a one of the deductive methodologies as analogous to the fault tree analysis (FTA) for determining the root causes of the hazard. This paper evaluates the advantages and disadvantages of STPA and CTA in the SOTIF perspective.