Hierarchical deterioration prediction for risk-based expressway maintenance

As the main artery supporting economic activity in Japan, the Metropolitan Expressway comprises many structures that have been in use for a long time since their construction. Due to the challenging operating environment, which includes a higher proportion of heavy vehicles compared to other road bridges, deterioration is remarkably advanced. Maintenance costs have continued to rise, while structural safety must still be ensured within limited budgets and personnel. In recent years, research on deterioration prediction using machine learning and using big data accumulated from repeated inspections has been progressing to address these challenges. This paper aims to achieve risk-based maintenance and management that considers the uncertainty observed in actual deterioration phenomena and proposes a ‘hierarchical deterioration prediction framework’.