Investigating accident causation through information network modelling

Management of risk in complex domains such as aviation relies heavily on post-event investigations, requiring complex approaches to fully understand the integration of multi-causal, multi-agent and multi-linear accident sequences. The Event Analysis of Systemic Teamwork methodology (EAST; Stanton et al. 200837. Stanton , N. A. and Baber , C. 2008 . Modelling of human alarm handling response times: a case study of the Ladbroke Grove rail accident in the UK . Ergonomics , 51 ( 4 ) : 423 – 440 .
[Taylor & Francis Online], [PubMed], [Web of Science ®]
View all references) offers such an approach based on network models. In this paper, we apply EAST to a well-known aviation accident case study, highlighting communication between agents as a central theme and investigating the potential for finding agents who were key to the accident. Ultimately, this work aims to develop a new model based on distributed situation awareness (DSA) to demonstrate that the risk inherent in a complex system is dependent on the information flowing within it. By identifying key agents and information elements, we can propose proactive design strategies to optimize the flow of information and help work towards avoiding aviation accidents.

Statement of Relevance: This paper introduces a novel application of an holistic methodology for understanding aviation accidents. Furthermore, it introduces an ongoing project developing a nonlinear and prospective method that centralises distributed situation awareness and communication as themes. The relevance of findings are discussed in the context of current ergonomic and aviation issues of design, training and human–system interaction.

Keywords

• aviation accident,
• communication,
• networks,
• distributed situation awareness,
• complex systems