Integrating Natural and Social Indicators for Enhanced Insurance Underwriting Strategies

Abstract

Facing extreme weather's growing impact, this study introduces a novel decision-making framework for the insurance sector. We use the Entropy Technique for Order of Preference by Similarity to the Ideal Solution (EW-TOPSIS) Model and Logistic Regression to assess underwriting risks and probabilities, underpinned by a Multi-Objective Planning Model that aims to optimize revenue and minimize churn. Case studies in Kumamoto Prefecture, Japan, and Texas, USA—areas affected by earthquakes and hurricanes—demonstrate the model's application. In 2022, it advised against underwriting in Kumamoto with profits of $1.325 billion and recommended a cautious strategy in Texas, yielding $3.544 billion. These results highlight the framework's utility in enhancing insurance strategies in extreme weather-prone areas for sustainability and profit.