Dynamic grey platform for efficient forecasting management
Dynamic grey platform for efficient forecasting management
复制标题
DOI：
10.1016/j.jcss.2014.12.011
复制
发表时间：
2015-09
期刊：
影响因子：
--
通讯作者：
Chen-Fang Tsai
中科院分区：
文献类型：
--
作者：
Chen-Fang Tsai
在本文中，我们提出了一个动态灰色平台，以修改传统的算法，通过应用两个新的预测算法预测管理。该平台集成了一个灰色模型（GM）与指数加权移动平均EWMA控制器称为EGM模型。EGM模型试图提高预测精度和效率。EGM模型的预测误差最小化，通过应用动态遗传算法（DGA）。DGA的贡献主要来自于它的两个特点：（1）遗传参数优化的交叉和变异率控制器;（2）EGM背景值优化的变量控制器。六个基准数据集已被用于模拟，以评估我们提出的模型的有效性。实验结果显示，较好的预测精度降低了台湾的绿色国内生产总值（GDP）的成本。
In this paper, we propose a dynamic grey platform to modify the traditional algorithms by applying two new prediction algorithms for forecasting management. The proposed platform integrates a grey model (GM) with an exponentially weighted moving average EWMA controller known as the EGM model. The EGM model attempts to improve the forecast accuracy and efficiency. The prediction error of the EGM model is minimized by applying a dynamic genetic algorithm (DGA). The contributions of the DGA are essentially from its two features: (1) the crossover and mutation rate controller of GA parameter optimization; and (2) the variable controller of EGM background value optimization. Six benchmarking data sets have been used in simulation to evaluate the effectiveness of our proposed model. The experimental results reveal that the better prediction accuracy reduces the cost of Taiwan's green gross domestic product (GDP).