In order to analyze the human reliability of the underwater man-machine cooperative search and rescue(SAR)op-eration and identify the key influencing factors,the influencing factors of human reliability were extracted according to the un-derwater SAR training practice and relevant research conclusions.An influencing factors system of human reliability was es-tablished,and the correlation between the influencing factors was qualitatively analyzed by the interpretative structural model.The interpretative structural model of the influencing factors was converted into the Bayesian network topology,and the Bayes-ian network for human reliability analysis regarding to three modes of operation,namely diver operation,remote operated vehi-cle(ROV)operation,man-machine cooperative operation was constructed,respectively.Through the reasoning analysis on three Bayesian network,the key factors affecting human reliability and their dependence were revealed,and the change in hu-man reliability under the three operation modes was quantitatively analyzed.The results show that the"misunderstanding in-tention","poor communication","information asymmetry","insufficient operation accuracy","loss of rescue capability"and other influencing factors play key roles in the man-machine cooperative SAR operation.When these factors are well con-trolled,the human reliability of the man-machine cooperative SAR operation will outperform the single ROV or diver inde-pendent rescue.The research results can provide reference for human reliability analysis and improvement of underwater SAR operations.
underwater search and rescueman-machine cooperationhuman reliabilityinterpretive structural modelBayesian network convergence
NETL
NSTL
万方数据
