Date of Award

Spring 5-2022

Document Type

Thesis

Degree Name

Master of Engineering (ME)

Department

Civil Engineering

First Advisor

Elizabeth Matthews

Abstract

Current practice of flood loss prediction presents limitations in accurately predicting building flood losses at multiple scales. While whole-building estimates can more accurately predict high-level losses (i.e., large groups of buildings), a significant analysis error is revealed with small-scale (i.e., individual, or small groups of buildings) investigation. This research presents a robust, data driven, building damage model seeking to elucidate a more fundamental understanding of flood damage of material components commonly used in residential construction. The framework of the model is based on a component-level damage database composed of data collected from experimental analysis. Structures with standard residential construction materials were built and incrementally flooded for short periods of time. The materials were assessed to determine the level of damage inflicted by the simulated flood events and catalogued based on material restorability. The restorability was determined through indicators such as moisture intrusion, corrosion, and mold contamination. The framework for the flood loss prediction model will be designed to incorporate damage uncertainty and be capable of analysis at multiple scales. This study not only provides a fundamental understanding of material damage, but also develops a more effective modeling tool of building community resilience through flood risk analysis and hazard mitigation planning.

Download

Share

COinS