Research on SHN Model Adjustment Method Based on K-Means Clustering and Gray Prediction Algorithm

In recent years, the frequency of extreme weather events has filled the insurance industry with uncertainty. In this paper, we use a model to quantify the underwriting conditions so that we can select appropriate areas for underwriting. First, we develop the Society-Human-Natural (SHN) assessment model for evaluating the underwriting safety level of a location. Based on data collected from 100 countries, nine secondary indicators were selected from social, human and natural aspects. Then, we combined Analytic Hierarchy Process (AHP), Criteria Importance Through Intercrieria Correlation (CRITIC) and Entropy Weight Method (EWM) to calculate the weights of each indicator and the USI (Underwriting Safety Index), and categorized them into three clusters by K-means clustering algorithm. In addition, considering the timeliness of insurance, we use the GM(1,1) model to predict the trend of USI. Finally, we selected the UK and Singapore to apply the model and calculated the USI to draw underwriting conclusions. Next, we adapt the SHN model considering the future situation of the real estate industry. First, we propose some strategies to improve the data and recalculate the corresponding weights. Finally, we recalculate the scores of the sample countries based on the adjusted SHN model and draw conclusions.