An Explorative Study On Risk Early-warning System Of Fund In Basic Medical Insurance System For Urban Working People

Objectives To design a scientific,weighable and operable Risk Early-warning System Of Fund in basic medical insurance system for urban working people. It can be used to measure and supervise the risk of fund , and give risk alarm when necessary, in order to give useful information to social security department and insure the fund safe.Methods The Risk Early-warning System Of Fund includes three hard-core modules: Early-warning Index System,Index Prediction Model and Risk Discriminant Model. This research designs Early-warning Index System with Delphi method and fuzzy AHP approach , frames Index Prediction Model with BP Neural Network, and contrives Risk Discriminant Model with Fuzzy BP Neural Network.Results Advised by 32 experts, Risk Early-warning Index System of fund comprises four indexes in Objectives and nineteen indexes in Alternatives. 80% index values of target quarter can be predicted very approximately with Index Prediction Model. The risk of fund in target quarter can be estimated accurately with Risk Discriminant Model.Conclusions Using BP Neural Network and some other methods to design Risk Early-warning System, can resolve the early-warning issue of fund in basic medical insurance system for urban working people and can do outstandingly.