A comparative inquiry into supply chain performance appraisal based on Support Vector Machine and neural network

This paper focuses on solving the practical problem of the supply chain performance appraisal. To improve the original evaluation methods, it constructs a new index system from a new angle of view based on survey and the existing outcomes. At the same time, it proposes a new theoretical evaluation model of supply chain performance appraisal based on support vector machine. Compared with the neural network method, the new model can overcome the disadvantages of the inherent instability, local minimum, slow convergence and poor ability of generalizing of the traditional methods. In addition, an empirical study has been conducted and the results show that the model based on support vector machine is effective and has more stable results, higher accuracy and better ability of generalizing than that of the neural network. Ultimately, it provides an effective method of supply chain performance appraisal for the managers in practical application.