This paper outlines the results of precursor analysis from 18 CoMET and Nova railways, all but two of which are metro (subway) operations, the other two being commuter railways. One of the commuter railways has mixed traffic including freight and long distance passenger trains, while four of the metro railways share lines with longer distance railways.
The paper analyses precursors, top events, injuries and deaths as well as safety maturity and their relationships both to each other and to incidents and accidents for the set of the major global metro railways. The 27 precursors analysed, for the period 2002-2009, fall into six categories: human performance; technical failures; passengers: fires; malicious action and management action.
To assess safety maturity, a model was developed based upon literature and metro experience. A questionnaire assessed the following safety maturity criteria for each metro: publication of safety reports; periodicity of safety monitoring; prioritisation of safety related incidents; efforts to mitigate risks and avoid incidents; description of safety procedures and monitoring of residual risks. The responses of each metro were scored to determine its aggregate safety maturity level.
Statistical analysis indicates a positive correlation between injuries and top events as well as between injuries and precursors. The safety maturity score is correlated with injuries but not with precursors, top events or fatalities. Using these results, this paper proposes a number of measures to improve metro railways safety in each precursor category.