A Bayesian Network Augmentation of JARUS SORA 2.5 for Probabilistic UAS Operational Risk Assessment: A Proof-of-Concept Study

This proof-of-concept work demonstrates a Bayesian Network (BN) augmentation of the Joint Authorities for Rulemaking on Unmanned Systems (JARUS) Specific Operations Risk Assessment (SORA) 2.5. The proposed architecture maintains the deterministic decision structure of SORA while introducing BN only where operational uncertainty exists. Mission characteristics and observable operational conditions are used as evidence, whereas uncertain factors, such as the risk of ground and air disturbances, are represented probabilistically. Consequently, the Bayesian Network enhances the assessment by quantifying uncertainty while maintaining the official SORA outputs, including the Ground Risk Class (GRC), residual Air Risk Class (ARC), and Specific Assurance and Integrity Level (SAIL). An airframe inspection mission at Riga Airport is used to demonstrate the applicability of the proposed framework and to compare its outputs with those of the SORA 2.5. The results indicate that BN provides an enhanced representation of operational uncertainty and causal relationships while preserving the deterministic structure of SORA. The framework is intended as a probabilistic decision-support tool rather than a replacement for regulatory assessment; therefore, this article does not estimate the probability of regulatory approval or claim predictive superiority over SORA. The contribution is a transparent model specification and a proof-of-concept foundation for future validation using empirical operational data and multiple mission scenarios.