A reliability estimation method is proposed to estimate the limit state exceeding probability of existing structures considering observation data by Sequential Monte Carlo Simulation (SMCS). Accuracy estimation is one of the important issues when MC approach is adopted. Specifically SMCS approach has a problem known as degeneracy. Effective sample size is introduced to estimate the accuracy of estimated limit state exceeding probabilities. Coefficient of variance of estimated probability is predicted based on the effective sample size ratio of the updated model. The proposed method is demonstrated through a numerical example of reliability analysis on deteriorating RC structures. © 2011 Taylor & Francis Group, London.