No description available.

The investigation and modelling of aviation accident causation is dominated by linear models. Aviation is, however, a complex system and as such suffers from being artificially manipulated into non-complex models and methods. This thesis addresses this issue by developing a new approach to investigating aviation accident causation through information networks. These networks centralise communication and the flow of information as key indicators of a system‟s health and risk. The holistic approach focuses on the system itself rather than any individual event. The activity and communication of constituent elements, both human and non-human agents, within that system is identified and highlights areas of system failure. The model offers many potential developments and some key areas are studied in this research. Through the centralisation of barriers and information nodes the method can be applied to almost any situation. The application of Bayesian mathematics to historical data populations provides scope for studying error migration and barrier manipulation. The thesis also provides application of these predictions to a flight simulator study in an attempt of validation. Beyond this the thesis also discusses the applicability of the approach to industry. Through working with a legacy airline the methods discussed are used as the basis for a new and forward-thinking safety management system. This holistic approach focuses on the system environment, the activity that takes place within it, the strategies used to conduct this activity, the way in which the constituent parts of the system (both human and non-human) interact and the behaviour required. Each stage of this thesis identifies and expands upon the potential of the information network approach maintaining firm focus on the overall health of a system. It is contended that through the further development and application of this approach, understanding of aviation risk can be improved.