中小企业在我国的经济发展中一直扮演着重要角色，以其庞大的数量与澎湃的动力推动着中国经济的快速稳定发展。但是，中小企业的融资问题却成为了制约其发展的一个阻碍，中小企业普遍规模较小、综合实力较弱、缺少融资相关体系。而供应链金融模式的出现，有效解决了中小企业的融资难题，同时也让中小企业能在更短时间内以更低的融资成本获得融资。然而，供应链金融模式也伴随着一系列的相关风险，这无疑会对融资的顺利进行以及银行与企业之间的合作产生影响。所以，当务之急便是建立起一套有效的供应链金融风险评估预测体系，并相对应地提出能够付诸实践的风险防范措施。首先，本文基于对供应链金融模式及相关风险的深入研究，确定了用于评估的各级指标，建立起总体的风险指标体系。之后，本文创新性地对定性指标进行量化处理、对原始指标采取因子分析将其浓缩为少数主因子，建立了基于因子分析的风险评估指标体系，以此增加指标体系的实际应用价值，并为后文风险预测模型的构建提供便利。接着，本文在前人研究的基础上，寻求构建预测精度及速度更优的供应链金融风险预测模型。本文基于因子分析的结果，分别使用基于变量筛选的Logistic回归与GA-BP神经网络来构建风险预测模型。在对它们进行全方面对比后，本文选取了基于GA-BP神经网络的供应链金融风险预测模型，能用以预测我国各中小企业的供应链金融风险，为商业银行及监管机构提供参考。最后，本文对GA-BP神经网络中各输入的显著水平进行排序，选取其中显著水平较高的作为关键主因子，并根据各关键主因子提出供应链金融风险的防范措施。此外，本文也结合供应链金融风险应对实例，来对提出的风险防范措施进行验证说明。本文构建了一套完整的基于GA-BP神经网络的供应链金融风险评估预测体系，并根据此体系结合实例提出了风险的防范措施，能够应用于实际，为参与供应链金融融资的我国各中小企业、商业银行以及监管机构提供参考。
Small and medium-sized enterprises (SMEs) have been playing an important role in China‘s economic development, promoting the rapid and stable development of China‘s economy with their huge number and surging power. However, the financing problem of small and medium-sized enterprises has become an obstacle to their development. Small and medium-sized enterprises are generally small in scale, weak in comprehensive strength and lack of financing related systems. The emergence of supply chain financial model has effectively solved the financing problem of small and medium-sized enterprises, and at the same time, it has enabled small and medium-sized enterprises to obtain financing at a lower financing cost in a shorter time.However, the supply chain financial model is also accompanied by a series of related risks, which will undoubtedly have an impact on the smooth financing and cooperation between banks and enterprises. Therefore, the urgent task is to establish an effective evaluation and prediction system of supply chain financial risks, and correspondingly put forward risk prevention measures that can be put into practice.Firstly, based on the in-depth study of supply chain financial model and related risks, this paper determines the indicators at all levels for evaluation and establishes the overall risk indicator system. Then, this paper innovatively quantifies the qualitative indicators, condenses the original indicators into a few main factors by factor analysis, and establishes a risk assessment index system based on factor analysis, so as to increase the practical application value of the index system and provide convenience for the construction of the risk prediction model in the following.Then, on the basis of previous studies, this paper seeks to build a supply chain financial risk prediction model with better prediction accuracy and speed. Based on the results of factor analysis, this paper uses Logistic regression based on variable screening and GA-BP neural network to construct risk prediction models. After comparing them in all aspects, this paper selects a supply chain financial risk prediction model based on GA-BP neural network, which can be used to predict the supply chain financial risks of small and medium-sized enterprises in China and provide reference for commercial banks and regulatory agencies.Finally, this paper sorts the significant levels of inputs in GA-BP neural network, selects the higher significant level as the key principal factor, and puts forward the preventive measures of supply chain financial risks according to each key principal factor. In addition, this paper also combined with the supply chain financial risk response examples to verify the proposed risk prevention measures.In this paper, a complete set of risk assessment and prediction system of supply chain finance based on GA-BP neural network is constructed, and according to this system, the risk prevention measures are put forward with examples, which can be applied to practice and provide reference for small and medium-sized enterprises, commercial banks and regulatory agencies involved in supply chain finance in China.
