Fuzzy-logic framework for updating the seismic fragility of deteriorating bridges via visual inspections

Abstract

Seismic evaluations of bridges are critical to ensure the post-earthquake integrity of transportation networks; however, systematic evaluations are costly. Hence, bridge management systems have developed screening policies to identify the most seismically vulnerable bridges. Nevertheless, such policies are typically qualitative and may not easily capture the impact of bridge aging or the uncertainty in material and structural parameters, leading to potentially underrated vulnerability estimates. Thus, a new quantitative-yet-practical framework is proposed to screen bridges given the fragilities of deteriorating critical components, namely bearings and columns. The framework uses fuzzy-logic to integrate deterioration data from routine visual inspections into fragility analysis. Then, parameterized fuzzy-fragility models are developed for columns and bearings given their deterioration level, bridge design parameters, and seismic intensity. These models are utilized to develop a seismic vulnerability ranking system to set bridge priorities for detailed seismic investigation. The analysis revealed that bridge vulnerability ranking could significantly increase with deterioration.