For statistical deterioration forecasting, many methods (e.g., Markov deterioration hazard model) have been proposed to estimate Markov transition probabilities using inspection databases with inhomogeneous inspection intervals, and there are many practical applications. These methods, however, require a high level of expertise from analysts, and there are technical obstacles to their use by practitioners. On the other hand, the counting estimation method with inspection database is conceptually understandable, but it is difficult to apply to the case where inspection intervals are inhomogeneous. In this study, an inspection database with inhomogeneous inspection intervals are pseudo-reconstructed so that the inspection intervals are homogeneous, and the counting estimation method is applied to the reconstructed database. Lastly, the usefulness of the proposed method is verified using an inspection database of real bridges, and the relevance of the proposed method to existing methods and the applicability of the proposed method in practice are discussed.
View full abstract