ABSTRACT
This paper presents a comparative study of the performance of different Monte Carlo Simulation methods in the computation of rare event probabilities in Reliability Theory. We evaluate the performance of 4 well known Markov Chain Monte Carlo methods (MCMC), namely Metropolis-Hasting (MH), Hamiltonian or Hybrid Monte Carlo (HYBRID), Delayed Rejection and Adaptive Metropolis (DRAM), and Differential Evolution Adaptive Metropolis (DREAM), for computing the Probability of Failure using the Reliability Theory framework. We also compared the results of simulations with an approximate analytical method called First Order Reliability Method (FORM). The study shows that while both HYBRID and DREAM produce more accurate results, contrary to intuition, HYBRID method was very slow in performance.
- Faming Liang, Chuanhai Liu, and Raymond J. C. 2010. Advanced Markov Chain Monte Carlo Methods: Learning from Past Samples, John Wiley & Sons, Ltd, 2010.Google Scholar
- Jasper A. Vrugt.,C. J. F. ter Braak., Diks C. G. H., and Bruce A. Robinson. 2009. "Accelerating Markov Chain Monte Carlo Simulation by Differential Evolution with Self-Adaptive Randomized Subspce Sampling," International Journal of Nonlinear Science and Numerical Simulation, vol 10, no. 3, pp. 271--288, 2009.Google Scholar
- Haario, H., Laine, M., Mira, A., and Sksman, E. 2006. "DRAM: Efficient adaptive MCMC," Statistics and Computing, vol. 16, no. 4, pp. 339--354, 2006. Google ScholarDigital Library
- Muhammed, A., and Speck, J. B. 2002. "Probabilistic Remnant Life Assessment of Corroding Pipelines Within A Risk-Based Framework," Paper presented at ASRANet International Colloquium 2002, 8--10 July 2002, University of Glasgow, Scotland.Google Scholar
- Stephens, M., and Van Roodselaar, A. 2009. "Developments in Reliability-Based Corrosion Management and the Significance of In-Line Inspection Uncertainties," 17th JTM Milan, 2009.Google Scholar
- Wu, Y-T., Hsaio, C. P., and van Roodselaar, A. 2009. "A Risk-Based Maintenance Optimization Methodology for Pipeline with Corrosion Defects," Proceedings of the ICOSSAR 2009.Google Scholar
- Wu, Y-T., and Shiao, M. 2009. "Sampling-Based Fast Probability Analyzer (FPA) for Reliability-Based Maintenance Optimization," Safety, Reliability and Risk of Structures, Infrastructures and Engineering Systems, Proceedings of the 10th International conference, Osaka, Japan, 2009Google Scholar
- Zhao, Yan-Gang, and Tetsuro Ono. 1999. "A general procedure for first/second-order reliability method (form/sorm)," Structural Safety, vol 21, no. 2, pp. 95--112, 1999.Google ScholarCross Ref
Index Terms
- A comparative study of Monte Carlo methods to compute rare event probabilities of failure in reliability models
Recommendations
Comparison of Monte Carlo and Quasi Monte Carlo Sampling Methods in High Dimensional Model Representation
SIMUL '09: Proceedings of the 2009 First International Conference on Advances in System SimulationA number of new techniques which improve the efficiency of Random Sampling-High Dimensional Model Representation (RS-HDMR) is presented. Comparison shows that Quasi Monte Carlo based HDMR (QRS-HDRM) significantly outperforms RS-HDMR. RS/QRS-HDRM based ...
Comments