Research Of Risk Measurement Model Based On Extended CVaR And Algorithm On Road Traffic Safety

With the rapid development of modern road traffic,road traffic safety problem is becoming the most difficult problem which threats countries all over the world.Under this background,efficacious methods to measure and control risk on road traffic safety are needed.It is important for the risk prevention and control on road traffic safety.This dissertation has introduced CVaR model to road traffic safety which is a tool for risk measurement in financial sector,and has extended CVaR model on account of the feature of road traffic safety.The extended CVaR has been used to build road traffic safety risk measurement multi-objective optimization model in the end.The main content is as follows:In chapter 1,it mainly summarizes the research of theory and applications on the related methods used in this thesis from the fields of road traffic safety risk,CVaR model and multi-objective particle swarm optimization algorithm.In chapter 2,it mainly gives the method of how to extract risk index of road traffic safety.This chapter summarizes the character of risk and analyses the feature of road traffic safety.And use Principal Component Analysis method to extract the most important index of road traffic safety risk in the end.In chapter 3,the risk measurement criteria CVaR model which is wildly used in financial sector is introduced in road traffic safety field.Base on the different features between road traffic safety risk and financial risk,this thesis expands loss function with one dimension in traditional CVaR to loss function with two dimensions.And then,build road traffic safety risk measurement multi-objective optimization model based on the expanded CVaR.A series of theorems with related proof are given to assure the comprehensive on theory of CVaR with two dimension loss function.In chapter 4,a new improved particle swarm optimization algorithm is proposed to solve the multi-objective optimization problems based on the extended CVaR model.A series of test functions are used to testify the performance of this improved algorithm,and the experimental results illustrate the outstanding performance advantage of the algorithm proposed in this thesis.In chapter 5,make use of the factual data of road traffic accident in our country using the risk measure method proposed in this thesis to measure the road traffic safety risk in our country at present.According to the results,the optimal control combination of road traffic safety risk in our country is given,and suggestion is made on how to prevent and control the road traffic safety risk in our country.In chapter 6,it introduces the main research contents,research results and innovation points;it also discusses the future research work.