THE CLASSIFICATION AND ANALYSIS OF RAILWAY INCIDENT REPORTS

This paper describes the method and some example results for a human factors review of railway incident reports. The analysis has initially been based on the TRACEr classification methodology (Shorrock, 2006) and developments of the approach by Nottingham University in the rail industry (RSSB, 2004). The paper will highlight how the practical application of the methodology in the rail context has led to changes in the classification approach and discusses the types of additional information which are required for the approach to work in the railway context. Examples of issues which have arisen in relation to classification are: the requirement to provide detailed task descriptions which place errors in context; a requirement to differentiate between performance shaping factors and errors; and a requirement to capture incident factors related to post-event management as well as those which are contributory or causal to events.