Comprehensive Evaluation on the Sustainable Development of Industrial Enterprises Using Artificial Neural Networks (ID: 6-058)

摘要

The sustainable development of industrial enterprises (SDIE) is in line with the scientific concept of development, and can obviously improve the competitive capacity of industrial enterprises. The evaluation of SDIE is a multivariate, comprehensive, nonlinear, tedious and difficult work. The artificial neural network was applied to evaluate SDIE. Based on the characteristics of SDIE, the used assessment indexes and their evaluation standards were used to generate efficient samples used for learning, verifying and testing the neural network model. According to the basic principles and steps in modeling and escaping from the local minimum, over-training and over-fitting, a robust and reliable artificial neural network-based (ANN-based) model for evaluating SDIE was established in this paper. Comparison with the traditional evaluation methods such as AHP (Analytic Hierarchy Processing), the ANN-based model, possessing accurate and objective, will be suitably applied to evaluate SDIE. Furthermore, ANN technique overcomes the shortcomings such as determining the weights subjectively, subordination functions of variables. Case study shows that the ANN-based model is powerful and effective in comprehensive evaluation of SDIE. For the ten indexes, the percentage of new products is the most important factor and the average utilization ratio of energy sources and gross sales the least.