Abstract: | This paper presents an analysis of accident precursors or ‘near misses’ (e.g. the space shuttle flights with damaged O-rings prior to the Challenger disaster). The interpretation of such events is often problematic, since the ambiguous nature of the evidence in such cases makes them subject to widely divergent interpretations. Simple Bayesian analysis provides a resolution of this problem, showing that an accident precursor almost always justifies increased rather than decreased estimates of overall accident frequencies. |