A study on the statistical deterioration prediction based on inspection data

Abstract

In this study, the deterioration prediction analysis is made by using the soundness evaluation value of each member in the inspection data that has been recorded in Ishikawa Prefecture. Because the inspection results contain inappropriate data for using in deterioration prediction, it is necessary to select data. A method is shown to identify the period in the deterioration progress and select the data to be used in the deterioration prediction by applying the piecewise linear regression. Many research and practical trials related to deterioration prediction by using the Markov deterioration hazard model are reported. Based on the current stored data, the deterioration prediction of Markov deterioration hazard model is compared with that of regression model.