2026 Volume 5 Issue 1 Pages 453-460
The authors have previously proposed a method for evaluating the residual structural performance of piers using deterioration assessment results, utilizing general-purpose structural analysis programs and artificial intelligence. However, it is anticipated that sufficient information, such as drawings, may be lacking for aging piers. Efficient maintenance is still required for such structures. Therefore, we investigated sensitivity analysis of explanatory variables in residual structural performance evaluation and methods to improve prediction accuracy by introducing additional explanatory variables. These findings were verified on actual piers. The results demonstrated that, depending on the assumed loading conditions, damage can be predicted accurately even with a limited number of explanatory variables.
