Assessing the Probability of the Dynamic Capsizing of Vessels

Abstract

Current naval vessel design requirements for reduced signature may drive industry toward unconventional hull forms, such as the tumblehome, for some vessels. There is a critical need to develop a system-based approach to intact and damaged dynamic capsize assessment that addresses risk identification and mitigation strategies, such as design methodologies, onboard "smart systems" for operator guidance, real-time seaway monitoring for critical conditions, monitoring extreme seaway ship handling trainers, and probability-based tools for risk assessment. Development of a system-based approach will also improve ship safety and allow the development of unconventional ship designs that rely on a variety of systems, including inherent design characteristics for minimizing intact and damaged dynamic capsize risk. Avail-ability of new intact and damaged dynamic capsize design methods and risk mitigation technologies will benefit both industry and the US Navy by breaking through the barrier of empirical methodologies for stability assessment currently in use. In this paper, an innovative method for computing the probability of dynamic capsizing of vessels is provided. The computation can be based on numerical simulation data or experimental data. In this paper, the method is illustrated using numerical simulation data applied to a hypothetical vessel. The suggested method utilizes the time to capsizing as a primary random variable for assessing the resulting time-dependent probability. The method utilizes models from reliability analysis based on life data, including the Kaplan-Meier technique, and would enable engineers to examine the run duration of tests, plan future tests including test repetition needs, and interpolate and predict capsizing probabilities under operational conditions that are not tested. In addition, the method could offer a basis for developing a system-based approach to assessing intact and damaged stability, navigation guidance procedures, and future risk-based navigation systems that include the initial design of hull forms, navigation procedures including human factors, and stability criteria. The method developed in this paper is presented using numerical simulation results for one vessel for the purpose of illustration. The illustration of the method starts with computing the time-dependent capsizing probabilities under various operational conditions of speed, heading, and random wave characteristics, called operational cells. Then probability distributions were fitted to the data and used for prediction purposes to gain additional insights, make observations, and draw conclusions for additional work. The case study results also produced recommendations for future work.