Third M.I.T. Conference on Computational Fluid and Solid Mechanics June 14–17, 2005  

Application of Markov chain Monte Carlo method for structural reliability

Krzysztof Kolanek*
Institute of Fundamental Technological Research, Polish Academy of Sciences, Swietokrzyska 21 Str., 00–049 Warsaw, Poland

  Full Text
Markov chain Monte Carlo is a method for simulation from a probability density function which is known up to a normalizing constant. Application of this method for evaluation of a probability of failure is presented. Several ideas which facilitate the proposed approach are provided. An illustrative numerical example is given.

Keywords:  Reliability analysis; Markov chain Monte Carlo; Normalizing constant; Exponential tilting; Importance sampling

* Tel.: +48 (22) 827 46 92; E-mail: