A comprehensive CREAM method for human reliability analysis of armored vehicle crews based on cognitive performance and operational environment

Given that human error is the primary factor causing combat task failures in armored vehicles, human reliability analysis (HRA) is very significant in enhancing human reliability and work efficiency for crews. To evaluate human reliability quantitatively and accurately, this study proposes a comprehensive cognitive reliability and error analysis method (CREAM). First, the weighting factors of different common performance conditions (CPCs) under uncertain conditions are derived by integrating the modified decision-making trial and evaluation laboratory-based analytic network process with linguistic D numbers. Second, considering the joint effects of cognitive performance and operational environment on crew behaviors, a cognitive performance adjustment coefficient is introduced to improve the conventional CREAM method. Third, group best-worst method and best-worst method based on nonlinear goal programming are used to determine the weighting factors of human intrinsic factors (HIFs). The results of the cross-platform combat task simulation show that the cumulative human error probability (HEP) of crews by this method is estimated as 26%, while the average HEP of the other HRA methods is approximately 24%. The HEP value has improved by 7% on average. The failure in judgment is the most critical contributor to human errors. Finally, according to the sensitivity analysis, the HEPs in different task processes with various CPCs and HIFs have significant differences (p < 0.01). The effect of the change in CPCs on the quantitative assessment of HEPs remains much steadier than that of the HIFs. The proposed method provides an effective method for the quantitative evaluation of human failure probabilities for crews in combat missions, which can decrease the security risk.