The Improved Fuzzy Classification Model on Disaster Loss

The aim of this paper is to solve the next two questions: (1) The quantitative assessment and classification of the disaster loss is a complex and difficult work. It often encounters the problem that different people will arbitrarily create different membership functions for the same issue, so the results of different model can not be compared. (2) The evaluation factors may have mutual relationships between each other, which will affect the results. The research methods of this paper are as follow: Firstly, this paper decomposes the related indexes based on orthogonalization theory, the aim of this step is to solve mutual relations between the factors leading to disaster losses; then confirms membership degree with the genetic algorithm (GA) method, getting the best membership of the factors; finally, this paper uses the fuzzy theory to recognize the damage degree of natural disaster, and evaluates the damage degree of mud-rock flow in Bijie area Guizhou province. The result is that the grade of disaster loss of Bijie area is low. The conclusion shows that the improved fuzzy classification model on disaster loss effectively solves the arbitrary and relevant problem of factors.