Probabilistic selection of event scenarios

Abstract

An extreme or accidental event in a structural system is always associated with multiple parameters which are random and probabilistic by nature. Therefore, a huge number of possible scenarios must be relevant for the real event. For advanced structural safety studies that involve refined computations and physical model testing, it is essential to select a set of realistic scenarios which must represent all the possible events but with a limited number which is needed to minimize the computational costs. This chapter presents practical methods to select a set of realistic scenarios but with a limited number by making use of probabilistic sampling techniques. The selected scenarios shall be used for advanced structural safety studies associated with limit states and risks.