Abstract
The present study is an attempt to develop the concrete bridge rating expert system with machine learning employing the hierarchical neural network to implement an inference mechanism. This network enables the system to facilitate the refinement of knowledge base by using the back propagation method. In this study, the training set (teaching data) for machine learning is obtained by the inspection for actual in-serve bridges and questionnaire surveys on bridge experts. Furthermore, comparisons between the diagnostic results of bridge experts and the ones of this proposed system are presents so as to demonstrate the validity of the learning capability of this system.
