A Comparison of Three Systemic Accident Analysis Methods Using 46 SPAD (Signals Passed at Danger) Incidents

摘要

During the period 1996-2003 there were five fatal accidents on the UK railway network, three of which were Signals Passed at Danger (SPAD) events (Watford Junction, 1996; Southall, 1997; Ladbroke Grove, 1999). SPAD events vary in severity and whilst most are not fatal there is the potential to cause serious injuries to passengers and train staff and damage to railway infrastructure. This paper investigates how the current system accident analysis tool used within the railway, the Incident Factor Classification System (IFCS) identifies and analyses causal factors of SPAD events. To evaluate the effectiveness IFCS was used to analysis SPAD incident reports (n= 46) and the outputs were compared with two systemic accident analysis methods and relevant outputs (the Human Factors Analysis and Classification System - HFACS and AcciMaps). The initial reporting process proved to hinder all systemic accident analysis methods in the extraction of causal factors. However, once extracted, all system accident analysis methods were successful in categorizing causal factors and demonstrated various outputs to illustrate the findings.