Analyzing reliability and sensitivity to reliability of rescue attachment with multiple failure modes

To improve rescue efficiency, a multi-function rescue attachment with tonging, shearing and grasping functions was designed by the authors in a previous study. In this paper, the reliability and reliability sensitivity of this rescue attachment are researched. Three rescue attachment failure modes are analyzed and limit state equations are established for the three failure modes. To consider the complex and nonlinear correlations among these failure modes, a method to analyze the reliability and sensitivity to reliability of the failure modes based on a hybrid copula model is proposed. The Clayton copula, Gumbel copula and Frank copula, which exhibit upper tail dependence, lower tail dependence, and upper and lower tail independence, respectively, are selected to construct the hybrid copula model to describe the complex correlations among multiple failure modes. The weight coefficients and parameters of the hybrid copula model are then estimated using the Bayesian model weighted average method and the minimum square deviation criterion. A method to calculate the sensitivity of the failure correlations to the reliability is then deduced using the moment method to improve the computational efficiency. Two examples show that the proposed method can improve computational efficiency significantly and the calculation results were consistent with those realized using the Monte Carlo (MC) method, but required fewer samples. Finally, the reliability and reliability sensitivity of the rescue attachment are analyzed using the proposed method and MC method. The effectiveness and efficiency of the proposed method are proved, and strategies to improve the rescue attachment's reliability are determined.