Performance evaluation model of cross border e-commerce supply chain based on LMBP feedback neural network

Abstract

In recent years, with the support of national policies, Cross Border E-Commerce (CBEC) has developed rapidly. This business model not only brings significant benefits to the national economy, but also has many unique challenges, especially at the level of supply chain management. Therefore, to enable CBEC enterprises to develop sustainable supply chain, this study discusses the performance evaluation model of supply chain and proposes a CBEC Supply Chain Performance Evaluation Model (CBECSC-EM) based on the Levenberg-Marquardt Backpropagation (LMBP) algorithm. This experiment constructs performance evaluation indicators for the supply chain of CBEC enterprises. On this basis, the LMBP algorithm is introduced, and improved in the experiment to make the overall performance of the evaluation model more scientific and reasonable. In the verification set, the maximum F1 values of LMBP, DEA, SBM, and BP are 98.46%, 93.78%, 87.29%, and 78.95%, respectively. The MAPE value of LMBP model is 0.102%, which is lower than the other three methods (0.282%, 0.343%, and 0.385%) selected in the experiment. The maximum standard deviation rates of importance and operability of the evaluation indexes are 0.1346 and 0.1405, respectively, and there is a significant consistency between the expert scores. Therefore, the LMBP algorithm has broad application prospects in supply chain performance evaluation of CBEC enterprises.