Yongpeng Zhao, Kouichi Takeya, Yuichi Ito, Eiichi Sasaki

Abstract: Considering the increasing number of aging infrastructure and the declining availability of skilled technicians in Japan, there is a growing demand for more efficient maintenance strategies. Leveraging information from existing inspection data presents a promising approach to address this challenge. This study proposes a novel model-interpretation-based framework that integrates multiple databases to quantitatively evaluate the effects of internal structural conditions and environmental factors on the corrosion of steel bridge main girders. By enhancing the interpretability of predictive models, the proposed framework provides actionable insights to support targeted data-driven maintenance planning. The proposed approach shows potential to be broadly applicable for the maintenance of various civil engineering structures, contributing to the development of more efficient inspection and maintenance programs, which can rapidly adapt to changing environmental conditions.

Keywords: Yongpeng Zhao, Kouichi Takeya, Yuichi Ito, Eiichi Sasaki

Date Published: December 22, 2025
DOI: 10.11159/ijci.2025.024

View Article