Date of Award

12-2016

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Mechanical and Civil Engineering

First Advisor

Jean-Paul Pinelli

Second Advisor

Nakin Suksawang

Third Advisor

Luis Daniel Otero

Fourth Advisor

Ashok Pandit

Abstract

Modeling hurricane events with models such as the Florida Public Hurricane Loss Model (FPHLM) is essential for insurance and reinsurance companies to estimate projected losses. Additionally, the models can be used by public officials to estimate and mitigate damages while also developing public safety policies to reduce the amount of damage and loss of life during a catastrophic event. The FPHLM combines hurricane event simulation and building vulnerability models to predict insurance portfolios losses. Before vulnerability functions are applied to insurance portfolios to determine total losses, these functions are weighted based upon building characteristics statistics representative of a given building population. To generate these statistics for the entire residential building stock in Florida, an extensive exposure study was carried on. This thesis focuses primarily on this exposure study. The exposure study encompasses the collection of raw exposure data and the subsequent generation of building characteristics statistics. A survey was conducted to obtain exposure data which was collected from a variety of resources including, but not limited to, private wind insurance portfolios provided by the Office of Insurance Regulations (OIR), National Flood Insurance Program (NFIP) portfolios, and county tax appraisers’ databases. This data was then formatted in a common fashion and statistics for building characteristics were generated independently for both Personal Residential (PR) and Commercial Residential (CR) buildings. While a data survey and a statistical analysis of the data is typical of exposure studies, other uses of the data were explored and additional studies were performed. These studies included refining the building statistics generated from a county to a zip code level. Also, a superimposition of the exposure databases from the different sources listed above was performed to identify which properties were insured for wind, flooding, wind and flooding, or neither. A framework was also developed for determining the elevation of structures within flood plains. In summary, this document contributes towards the exposure component of the FPHLM in which statistics for building characteristics were determined for PR and CR structures. The availability of a relatively complete exposure data set allowed for building trends to be identified and for other studies to be performed which could lead to future improvements in the FPHLM as a whole.

Download

Share

COinS