In order to study the risk management of supply chain, promote collaboration among node enterprises, and ensure the normal and good development of supply chain, a risk evaluation model of supply chain based on BP (Back Propagation) neural network (BPNN) was constructed to explore the existing factors affecting the risk evaluation of the supply chain and to build risk evaluation index system of supply chain. The results show that BPNN has unique advantages in solving such highly non-linear problems as evaluation of risks of the supply chain. The proposed method can be regarded as an effective technical means to determine a supply chain risk and provide current enterprises with effective decision support to carry out the risk management of the supply chain. As a result, the proposed model will provide a good reference for enterprises and other financial institutions to expand the financing business of supply chain and facilitate the evaluation of enterprise credit risk.