The purpose of a bridge maintenance strategy is to make effective decisions by evaluating current performance and predicting future conditions of the bridge. The social cost because of the rapid increase in the number of decrepit bridges. The current bridge maintenance system relies on traditional man-power-based methods, which determine the bridge performance by employing a material deterioration model, and thus shows uncertainty in predicting the bridge performance. In this study, a new type of performance degradation model is developed using the actual concrete deck condition index (or grade) data of the general bridge inspection history database (1995–2017) on the national road bridge of the bridge management system in Korea. The developed model uses the long short-term memory algorithm, which is a type of recurrent neural network, as well as layer normalization and label smoothing to improve the applicability of basic data. This model can express the discrete historical degradation indices in continuous form according to the service life. In addition, it enables the prediction of bridge performance by using only basic information about new and existing bridges.
This is an open access article distributed under the Creative Commons Attribution License
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited