Quantitative analysis of enterprise chain risk based on SVM algorithm and mathematical fuzzy set

Abstract

Under the guidance and practice of lean production, just in time production and other advanced theories, the relationship between enterprises is becoming more and more closely. In order to cope with more fierce market competition, manufacturing enterprises began to strengthen cooperation with partners in the supply chain, gather resources, improve competitiveness and jointly fight against competitors. In these decades, the competition among enterprises is gradually replaced by the competition among supply chains. In this paper, the author makes quantitative analysis of enterprise chain risk based on SVM algorithm and mathematical fuzzy set. Support vector machine (SVM) is a machine learning method, has strong generalization ability and accuracy. By analyzing dexterity affects the normal operation of the supply chain risk factors, we use simulated annealing –mathematical fuzzy of the risk evaluation, it indicates that the model in risk assessment is applicable through empirical research. According to the data obtained, the simulated annealing –support vector machine evaluation model were trained and tested; the explanation on the choice of kernel function of the process of construction of the evaluation model, the parameters of the model to determine some key problems.