Research On Credit Risk Of Supply Chain Finance In Automobile Industry

With the development of our national economy,SMEs are playing an increasingly important role.They play an irreplaceable role in innovating technology,increasing employment,stabilizing society and promoting economic growth.However,due to the small-scale operation of SMEs,the credit level is generally not high,the ability to resist risks is poor,poor transparency of information and so on,financing difficulties has always been an important issue facing it,and the emergence of supply chain finance provides a good way for SMEs to solve financing difficulties.Compared with the traditional direct financing to banks,the supply chain finance business examines the overall situation of the supply chain,which increases the difficulty of commercial banks in assessing the credit risk of financing enterprises.Based on this,this paper selects the auto parts enterprises as the research object,comprehensively analyzes the credit risk of supply chain finance,and conducts empirical research to build a credit risk evaluation model.Based on the two theoretical foundations of supply chain finance theory and risk management theory,this paper reviews the current research status of supply chain finance risk at home and abroad,combines it with the reality of the automotive industry,and analyzes the characteristics and causes of its credit risk.And referring to the research results of the mainstream rating agencies,financial institutions and experts and scholars,the evaluation indicators of the credit risk of auto parts enterprises in this article are determined as the six first-level indicators of financing companies,core companies,logistics companies,supply chain operations,credit supporting assets and industry conditions,and further subdivided into 16 secondary indexes and 35 tertiary indicators.Through the entropy method and correlation analysis,expert scoring method for the second screening of indicators.Finally,the credit risk evaluation index system of auto parts enterprises containing 5 quantitative indexes and 17 qualitative indexes is established.By comparing various credit risk assessment methods,the credit risk assessment method of this paper is identified as BP neural network.In addition,considering that the BP neural network can easily fall into the local optimum,this paper introduces the shuffled frog leaping algorithm to improve it.The final construction of the credit risk assessment model of auto parts enterprise based on BP neural network is a three-layer BP neural network model with 22 input layer nodes,8 hidden layer nodes,and 1 output layer nodes.The network learning rate is 0.01,the target error is 0.001,and the maximum number of trainings is 10000.Through a questionnaire survey,this paper collected 90 valid questionnaires and randomly divided them into 60 training samples and 30 validation samples.The simulation results show that the accuracy rate of the training samples is about 100%,the accuracy rate of the verification samples is 86.7%,the accuracy rate of all samples is 95.6%.The model determines that the accuracy of the financing enterprises with high credit risk is greater than that of the financing enterprises with low credit risk,and the error rate is significantly lower compared with the BP neural network model before the improvement.This proves the validity and accuracy of the established credit risk assessment model.This will have certain reference significance for financial institutions such as commercial banks to evaluate the credit risk of financing enterprises when carrying out supply chain finance business.