The Research Into Crisis Early Warning Of Supply Chain Energy Saving And Emission Reduction Based On FSVM

With the rapid development of economy, the competition between enterprises becomes increasingly intense, resulting in excessive consumption of resources and serious environment pollution. Under this background, our country put forward the concept of "beautiful China" to vigorously promote energy conservation and emissions reduction, and to build a resource thrift and environment-friendly society. Therefore, the enterprises must comprehensively consider economical, rational and socially and environmentally acceptable, technically feasible measures, to reduce energy consumption, pollutant emission; to achieve least negative effects on the overall supply chain environment and highest resource utilization rate of in the process of obtaining product raw material, design and manufacture, sales and transportation, usage and recycle in the process of supply chain. Based on the above situation, energy saving and emission reduction of supply chain is particularly important. Meanwhile, the globalization of supply chain intensifies, enterprises within this industry become closer to each other and the structure more complex, any linking problem will affect the next step of energy conservation and emissions reduction, further create crisis. Therefore, it is essential to conduct a research on early crisis warning of supply chain energy saving and emission reduction.Firstly, this paper is based on energy saving and emission reduction, supply chain risk management and crisis early-warning management theory these three aspects of the research status at home and abroad. This article also gives a brief introduction to Fuzzy Clustering (FCM) and Support Vector Machine(SVM) theory, which lays a solid theoretical foundation for further research.Secondly, referring to the research achievements of the financial crisis and following a five-step measure (clear warning—looking for the alert source— determine warning factor—forecasting warning degrees——remove warning) to establish the module of crisis early warning of supply chain energy saving and emission reduction. In looking for the alert source module, this paper fully analyze the process of supply chain of energy-saving emission reduction, thus concludes a source classification system of warning of supply chain energy saving and emission reduction’s crisis.Thirdly, the initial crisis early warning index system of the supply chain energy saving and emission reduction has been built up on the basis of literatures, government reports and enterprises surveys. After feasibility, importance, distinction evaluation of the experts, we are able to acquire the data that finally leads us to an all-around crisis early warning index system.Finally, data of this paper was collected from the enterprise questionnaire elaborately designed the crisis early warning index system that was previously built up. Then applying the Fuzzy Clustering (FCM) to classify,using Fuzzy Support Vector Machine (FSVM) for data segmentation,increasing the fuzzy membership of each sample, eventually. three-level Fuzzy Support Vector Machine based on FSVM is built.The empirical results show that,the FSVM-based model not only improves the speed and accuracy of the classification model, but also effectively solves the problem of the support vector machine can not be separated, this verify the feasibility of the model.