Derivation of deterioration prediction interval with stochastic Markov Chain model and Bayes estimation of individual deterioration characteristics of bridge

Abstract

This manuscript presents a new bridge deterioration prediction method with stochastic Markov Chain model, which uses the transition probability matrix whose elements are described as random variables. The model is able to develop a prediction interval of bridge deterioration while the conventional model is not able to. In addition, we have developed the method to consider the individual deterioration characteristic expressed as the inspection results by Bayesian theory. The prediction interval developed by the proposed model allows bridge owners to understand the deterioration in several ways, including safe-side prediction and average prediction. It is expected to provide a better insight into the estimation of maximum, average, and minimum maintenance cost to the bridge owners, which makes their decision making process more productive.