A Study of the Insurance Industry in Extreme Weather Based on ARIMA Models and EWM-TOPSIS

Abstract

This paper evaluates the underwriting approach of property insurance by comparing the total premium per person and affordability per person, using the ARIMA model to predict the underwriting decisions in Japan and Turkey for the next five years, combining factors such as population density and economic development on a smaller regional scale, and establishing a second model, the FRA. applying the theory of catastrophic factors, the Fuzzy Risk Assessment Model calculates the causal factor intensity risk values and vulnerability values to derive a more accurate integrated risk assessment value. Afterwards, the URA model is applied to the problem of community siting in Australia and the C-GIS model is established. By analysing factors such as terrain data, population distribution and average temperature to determine suitable locations for development, the URA model is added to analyse the risk level under extreme weather and generate satellite maps for real estate siting. Finally, buildings of cultural or community importance in the area are considered, and a DPB model is built to solve the risk scoring and protection method using EWM-TOPSIS, which is classified into three levels: relocation protection, restricted protection and normal protection.