Light Pollution Assessment Decision Model Based on Entropy Weight Topsis and Decision Tree

Based on recent global night scene data, this study explores the relationship between light pollution and factors such as population, economy, environment, and geographical location. By utilizing indicators such as EKC-GDP (regression-based GDP index), artificial light intensity (LP), population density, nearest distance from the coastline, and vegetation coverage, an evaluation model for light pollution level is established using the Entropy-TOPSIS method, and a decision tree classification model is constructed to categorize the study area into three categories: A, B, and C. Three strategies are developed based on characteristics of each category: the Dark City Strategy, the Green Space Strategy, and the Bright Light Strategy. Our model is not sensitive to the change of artificial light intensity, and can be applied to a variety of areas with different degrees of light pollution. We believe that the model is stable, and the three classification results and three governance strategies can be applied to the global scope.