Using modified grey forecasting models to forecast the growth trends of green materials

作者：Tsai Sang Bing^*; Lee Yu Cheng; Guo Jiann Jong

来源：Proceedings of the Institution of Mechanical Engineers - Part B: Journal of Engineering Manufacture , 2014, 228(6): 931-940.

DOI：10.1177/0954405413509079

• Forecasting
• grey system theory
• grey model(1,1)
• non-linear grey Bernoulli model(1,1)
• grey Verhulst model
• production research
• small data set
• green corporate social responsibility

The use of green materials reflects the notion that production materials should generate minimal environmental pollution through various efforts, including the design and development of low-pollution materials, innovation and improvement of low-pollution manufacturing processes, and recycling and reuse of materials. The development of green materials is an important part of corporate social responsibility. Companies need to use resources legitimately and have environmental protection responsibility. Forecasting the growth trends of green copper clad laminate material is crucial for manufacturers of printed circuit boards and green copper clad laminates. The main purpose of this study was to forecast the growth trends of green electronic materials. The industry sample investigated in this study only numbered 14. Because of the limited sample of historical data, the data distribution did not exhibit a normal distribution. Prediction methods for large data sets were not suitable for this study. Thus, three grey forecasting models (i.e. the grey model(1,1), non-linear grey Bernoulli model(1,1), and grey Verhulst model) were adopted for theoretical derivation and scientific verification. The results yielded by these methods were compared with the results of regression analysis to verify the forecasting accuracy and suitability of the three methods. The results indicated that for small data sets, the forecasting accuracy of the non-linear grey Bernoulli model(1,1) and grey Verhulst model was superior to that of the original grey model(1,1) as well as the regression analysis method.

• 出版日期2014-6
• 单位南开大学