Hierarchical Bayesian Estimation of Mixed Hazard Models

A variety of uncertainties affect the deterioration process of infrastructure. Deterioration rate is varying significantly according to the difference in the structural characteristics, use, and environmental conditions of infrastructures. The heterogeneity of structures causes the problem of overdispersion of the deterioration rate. In order to overcome the problem of overdispersion, the mixed Markov deterioration hazard model has been proposed considering the heterogeneity of deterioration rate among groups of infrastructures. In this study, it is assumed that the overdispersion depends on the heterogeneity of the deterioration rate among groups of infrastructures. Then, the mixed Markov deterioration hazard model that takes into account hierarchical heterogeneity is formulated, and a hierarchical Bayesian estimation method is proposed. The appropriateness of the proposed method is discussed through the empirical analysis of the visual inspection data of bridge slabs.