2016 Volume 72 Issue 2 Pages 191-210
In this study, the authors develop a deterioration prediction model with respect to deterioration management indicators by using continuous variables. Quite a few deterioration management indicators of infrastructure are observed as continuous values. It is possible to discretize these continuous values and predict deterioration with a Markov chain model, but in order to obtain more precise deterioration prediction results, it is desirable to develop a direct model of the time variation in continuous values. In this study, the authors formulate a continuous deterioration hazard model including characteristic variables that vary according to the structural characteristics and environmental conditions of infrastructure, and propose a Bayesian estimation method using the Markov Chain Monte Carlo method. Furthermore, the authors attempt to empirically analyze the applicability of the methodology proposed in this study.