Probabilistic Load Model for Ship-Bridge Collision Using Monte Carlo Simulation

In designing a bridge crossing navigable waterway, ship-bridge collision is the one of the critical extreme events. Due to its inherent uncertainties, probabilistic design approaches have been mainly adopted, in which the probability of failure or collapse of the whole bridge system should be estimated. Therefore, it is crucial that appropriate probabilistic models should be employed for collision loads and resistances of the piers. In this paper, a method is presented for developing probabilistic load models for ship-bridge collision. The difficulties arise from the fact that the resultant collision force, which is a function of other random variables such as speed of a ship, collision angle and position, etc. Furthermore, statistical data of actual ship-bridge collision events are very rare. Thus, a simulation based approach is proposed for the probabilistic load model for ship-bridge collision. The collision force model is derived from the combination of the probability of the collision event and the conditional probability distribution of the impact load of each collision case. In this approach, models are also incorporated for initial ship position, aberrance probability and human intervention stopping ship before collision. For the impact load,

uncertainties such as collision angles, velocities and the weight of colliding ship are also considered. The final probabilistiy distribution of collision load is constructed based on Monte Carlo simulation method. A numerical example of an actual cable-stayed bridge is presented for the demonstration of the proposed model.