Time-variant hull girder reliability considering spatial dependence of corrosion growth, geometric and material properties

The stochastic deterioration of ship structural elements is both time and spatially dependent. In this paper, a formulation of the time-variant ship hull girder reliability subjected to corrosion is proposed by taking into consideration the spatial variability of corrosion growth, geometric and material properties of structural elements. The joint probability distribution of the spatially dependent variables is constructed by the Gaussian, Gumbel, Clayton, and Frank copulas. Monte Carlo simulation is used to compute the probability of hull failure. A ship hull girder is employed to illustrate the application of the developed method. The impact of the degree of dependence and the selected copula on the estimated failure probability is investigated. The results indicate that this probability can differ assuming different levels of spatial dependence and different dependence structures of corrosion growth, geometric and material properties. In addition, a higher dependence of variables tends to produce higher probability of hull failure.