For a ship operating in ocean waves, ensuring safety and durability requires consideration of the statistical characteristics of the ocean environment over its service life. Among various structural responses, the vertical bending moment (VBM) is one of the most critical response modes to be addressed, as excessive VBM may lead to collapse of the hull girder. Long-term risk can be evaluated through long-term prediction (LTP), which aggregates multiple short-term predictions (STPs) of extreme responses under various sea states and operational conditions. Conventionally, the statistical characteristics of the ocean environment expected to be encountered over the operational period—e.g., 25 years in the case of a sailing ship—are used to predict long-term extreme responses, referred to as LTP [1]. However, as numerical models used for STPs increase in computational cost in order to account for inherent nonlinearities in the structural response, the direct application of Monte Carlo integration becomes computationally prohibitive. Consequently, the high computational burden associated with LTP hinders the practical incorporation of reliability-based design optimization (RBDO) for mitigating long-term risk. To address this issue, in this work, a new active learning methodology is proposed for the efficient execution of RBDO targeting the long-term mean outcrossing rate (MOR). The proposed RBDO framework incorporates Bayesian active learning (BAL) to efficiently estimate the MOR of VBM, thereby expediting the LTP process. Numerical demonstrations are presented to optimize the longitudinal weight distribution of a ship so as to mitigate the long-term MOR of VBM. By applying the proposed RBDO approach, the highly nonlinear characteristics of VBM arising from the hydroelastic response of a ship can be explicitly accounted for in the design process. Finally, the versatility of the present methodology is discussed in the context of RBDO for general marine structures. REFERENCES [1] L. Sagrilo, A. Naess, A. Doria, On the long-term response of marine structures, Applied Ocean Research, Vol.182, pp. 208–214, 2011.