Development of the Damage Attribution Chart Identifying the Wind-Susceptible Building Component by Incorporating Fault-Tree Analysis into the Component-based Vulnerability Modelling Against Severe Wind – A Case of the Philippine Public-School Building

Abstract

The component-based approach in vulnerability modelling can estimate the amount of damage per vulnerable building component. However, it cannot identify the source of failure. Hence, this study incorporates a fault-tree analysis into the vulnerability model in order to identify the susceptible component. Thereby developing a systematic framework that aids in directing where building strengthening efforts must be applied. The framework is applied to a one-story, three-bay public-school building, where the vulnerable building components identified and being monitored includes the roof fastener, roof cover, purlin connection, purlin section, roof-to-column connection, and the window panels. The building is subjected to wind pressures estimated using the computational fluid dynamics (CFD) modelled in ANSYS and calibrated with a wind tunnel test. These loadings together with the probabilistic resistance capacities determined for the identified vulnerable components are fed into the Monte-Carlo simulation with fault-tree algorithm to quantify and classify the damage. The failure simulation produces what this study coined as the damage attribution chart that shows which building component can the damages be attributed to. The resulting critical component for the case study considered is the roof-to-column connection. This was validated against a field survey conducted to a similar school building that was damaged by the Typhoon Odette, which shows similar damages as simulated. The study further investigates the effects of strengthening various building components which suggests that not all strengthening mitigations would lead to the reduction of the overall building damage.

Keywords: Damage Attribution Chart; Component-based Reliability Modelling; Wind Vulnerability Curves; Design Optimization and Retrofit.