Applying Probabilistic Risk Assessment to Safety Risk Analysis in Aviation

Background

Towards managing safety, the SSMT program has been involved in the design and development of an Integrated Safety Assessment Model (ISAM) for the analysis and assessment of risk in the NAS. Safety models are qualitative and/or quantitative representations of the system that assist in the identification and management of risk. The qualitative aspects of the model capture what must go wrong for a failure within the system to occur, and the quantitative aspects capture how probable or severe that failure is. The core of ISAM is an integrated pilot-controller safety model developed through adaptation of two European models, the Causal Model for Air Transport Safety (CATS) (Ale et al., 2009) and the EUROCONTROL Integrated Risk Picture (IRP) (Spouge & Perrin, 2006). The CATS safety model is a representation of world-wide historic accidents and incidents, and includes a Bayesian Belief Net (BBN) – based causal model. The ISAM model has adapted the basic designs of the IRP and CATS models for representation of accident and incident scenarios and consequences and continues to explore how improvements in the model structure and design such as those suggested by Hu, Zhang, Ma, & Liang (2011) and Cai et al. (2013), might affect model fidelity.