A Bayesian network version of TRACEr for human reliability analysis of the space launch mission

This study applied the Technique for Retrospective and Predictive Analysis of Cognitive Errors (TRACEr) to the analysis of the space launch mission. The analysis is consisted of the following two main stages: the identification of human errors in the space launch task context with the TRACEr and the quantification of human error probability (HEP) with a Bayesian network (BNN) based extension of the TRACEr. In the first analysis stage, thirty-seven mishap reports of the Chinese space launch activity are collected and analyzed. Based on the analysis results, the structure of the BNN is determined. In the second analysis stage, with incorporating the experts’ judgements, the HEP of the task in space launch activity is produced. Also, the sensitivity analysis has been conducted and the most significant influencing factors on human errors in space launch mission have been determined.