Analysis on the Utilization of Opinion of Bridge Inspection Results Using Topic Model for Maintenance and Management

Abstract

Utilization of collected and accumulated data of bridge inspection results is required for maintenance. Although the qualitative evaluation items of bridges have frequently been used in the past, this research focuses on the opinions of technical experts during bridge inspection. Topic models consist of probabilistic modeling of term frequency occurrences in documents, which can be used to classify documents containing opinions of bridge inspections. A topic model can be applied to the text data to detect the characteristics of bridge damage not recorded as database evaluation items. The topics corresponding to the characteristic damage condition of the bridge can then be organized. In this study, topics have been organized by high probability words about deterioration due to ASR (Alkali-Silica Reaction), which is a characteristic deterioration in the Hokuriku region. By focusing on the opinions from bridge inspections associated with high probability topics, early detection and preventative maintenance are possible. Although effective results were obtained, some topics were organized by the characteristic writing of the inspection company. This means that in order to accumulate the inspection data while maintaining quality, it is necessary to consider the method and writing style that was used in creating the contents of the opinion in the bridge inspection report.

 

This Paper was Originally Published in J. Soc. Mater. Sci., Japan 69 (2020) 197–203.