Research On Credit Risk Evaluation And Decision Of Small And Medium Equipment Manufacturing Enterprises Under Supply Chain Finance

The impact of the spread of the Corona Virus Disease 2019 in 2020 on the global economy has exceeded the 2008 financial crisis,and a large number of small and medium-sized enterprises need capital investment to resume operations.China’s Ministry of Industry and Information Technology has successively issued notices to promote the application of supply chain finance financing methods.At present,the annual accounts receivable of small and medium-sized enterprises is about 26 trillion yuan,and the total amount of accounts receivable financing of financial institutions and factoring companies is about 220 billion yuan,accounting for less than1%,and there is a huge financing space.The equipment manufacturing industry has typical supply chain characteristics,with a clear division of labor between upstream,midstream,and downstream,and is a key industry for supply chain finance.Affected by the “Matthew Effect” of financial supply,the overall financial support of the equipment manufacturing industry is uneven.The banking industry must comprehensively consider factors such as corporate credit rating,asset-liability status,and project risks.Large equipment manufacturing companies,as high-quality resources,are more in line with bank credit requirements,resulting in a relatively concentrated supply of financial services to several large state-owned equipment companies,and a "crowding out effect" on the financial supply of small and medium-sized enterprises,,There are financial service gaps for some small and medium-sized enterprises and emerging important technological fields,and the implementation of technological innovation in small and medium-sized equipment manufacturing enterprises is restricted,which has widened the gap in the development process of different types of equipment manufacturing enterprises.“Matthew Effect” is becoming more obvious,which will restrict the overall development of the equipment industry.Firstly,based on the perspective of the whole supply chain,this paper systematically analyzes the hierarchical and robust characteristics of the equipment manufacturing supply chain,and determines the credit risk index system including the qualification of small and medium-sized enterprises,the status of supply chain network,the qualification of supply chain cooperative enterprises,the asset status under financing and the macro environment.Secondly,the stochastic forest is optimized by genetic algorithm to build a credit risk evaluation model to measure the credit default risk of small and medium-sized equipment manufacturing enterprises in the supply chain finance mode.Finally,the supply chain adjustment factor is introduced,and the credit line model and loan pricing model are established to provide reference for the credit decision-making of small and medium-sized equipment manufacturing enterprises.This research puts forward innovative thinking on the credit risk assessment of supply chain finance financing model,and provides practical reference for solving the loan dilemma of small and medium equipment manufacturing enterprises under the supply chain finance model.The main work and conclusions of this research are as follows:(1)The credit evaluation index system of small and medium-sized equipment manufacturing enterprises with supply chain characteristics and significant differentiation of risk factors is constructed.This paper takes 940 equipment manufacturing industry samples of SME Board of Shenzhen Stock Exchange from 2014 to 2018 as the research object,and obtains the sample data through Tianyancha,enterprise official website,Eastmoney information,company annual report and other channels.Based on the characteristics of equipment manufacturing supply chain,such as high robustness,complete chain structure,obvious hierarchy and high degree of information sharing,two dimensions of transaction breadth and transaction depth are introduced to measure the status of supply chain network.The credit risk index system includes the qualification of small and medium-sized enterprises,the status of supply chain network,the qualification of supply chain cooperative enterprises,the asset status under financing and the macro environment.According to the partial correlation-variance analysis,64 indicators with redundant information and little impact on default status were deleted.According to the optimal principle of overall risk factor identification,the indicators that have a negative impact on the overall default identification ability are eliminated one by one.Finally,the credit risk evaluation index system of small and medium-sized equipment manufacturing enterprises under the supply chain finance mode is established,which contains 56 indexes.The empirical results show that the credit risk evaluation index system for small and medium-sized equipment manufacturing enterprises under supply chain finance established in this paper fully reveals the characteristics of the supply chain and equipment manufacturing industry,conforms to the “5C principle” generally recognized by the financial industry,and the accuracy of enterprise risk factor discrimination is as high as 90.53%.(2)The credit risk evaluation model of small and medium-sized equipment manufacturing enterprises in supply chain finance based on GA-RF is constructed.This paper uses the K-Means under-sampling model to cluster the large-class samples through the clustering method to form the same number of clusters as the small-class samples.A single sample is randomly selected from each cluster to form a balanced sample set with small class samples.Aiming at the problem that the existing grid search algorithm takes a long time and cannot reach the global optimal solution,this research introduces the genetic algorithm into the random forest for optimization,and changes the number and depth of the decision tree in a binary population.Through the parent multi-point crossover and the mutation generates high-quality progenies,and the optimal parameter combination is obtained through the standard iteration with the largest fitness value,that is,the highest classification accuracy.The existing research on the random forest grid search algorithm is improved,which has a large amount of calculation and low classification accuracy,and constructed a credit risk evaluation model for small and medium equipment manufacturing enterprises under supply chain finance.The empirical results show that the overall discrimination accuracy,precision rate,and recall rate under the supply chain finance model are better than the traditional loan model.The random forest model is higher than the other four models in overall discrimination accuracy and AUC value.Random forest is more suitable as a credit risk assessment model for SMEs.The improved GA-RF model has been improved in all discrimination standards,and the overall discrimination accuracy has reached 92.45%.(3)The multi-factor-oriented credit decision model under the supply chain finance model is constructed.The credit model comprehensively considers the three factors of maximum debt space,corporate growth capacity,and credit rating.Based on the customer’s comprehensive solvency,variables such as sustainable growth rate and credit rating adjustment coefficient are added to reflect corporate growth capacity and credit rating,and the repayment ability of enterprises is comprehensively evaluated.The loan pricing model comprehensively considers the optimal ratio of commercial banks’ revenue and costs,constructs a SME loan pricing target planning model with the largest interest and the smallest cost,and calculates the combination coefficient of the target programming function through Lagrange method,so as to provide the best scheme for commercial banks to price loans for small and medium-sized enterprises.On this basis,the optimal cut-off point score is further calculated.By analyzing the difference between the credit score and the cut-off point,the supply chain operation reward and punishment mechanism is introduced as the adjustment factor,and a certain proportion of preference and punishment are given to the credit line and loan pricing respectively.The bank loan decision-making model under the supply chain finance mode is constructed.The empirical results show that whether the loan period is 1,2,or 3 years,the supply chain loan interest rate is lower than the interest rate in the absence of preferential treatment.The better the overall credit performance of the supply chain,the more preferential loan interest rates can be obtained,and the loan interest rate of risky enterprises is significantly higher than that of risk-free enterprises.After adding the supply chain adjustment factor,most companies have increased their credit lines,and some companies have even achieved significant growth,but there are also a few companies with poor supply chain performance that have reduced their credit lines.