The concrete analyzing process uses the vertical contrasts of time and horizontal comparison of region as the main technique method . the forecasting course has used mathematical approach to make science decisions such as correlation recurrence and exponential smoothing etc 本论文主要采用了现状分析为基础,理论为指导的研究方式;具体分析过程采用时间纵向对比、区域横向比较的分析方法说明问题;预测过程采用了相关回归、指数平滑等数学方法科学决策。
The forecasts method including the forecast method of simple moving average , the forecast method of weighting moving average , the forecast method of single exponential smoothing , the forecast method of double exponential smoothing , the forecast method of multiplication model and the forecast method of monadic linear regression 预测方法包括简单移动平均法、加权移动平均法、一次指数平滑法、二次指数平滑法、乘法模型预测法和一元线性回归方程预测法。
According to the case that the macrocosmic system is nonlinear and lack of testing data , the improved forecasting methods are proposed such as fuzzy exponential smoothing forecasting , center approaching gray prediction and the local multiple regression fuzzy ( lmrf ) model based on takagi - sugeno fuzzy logical system . these improved methods are applied into the forecasting instances . the prediction accuracy of the stimulation result is testified and the improved forecasting methods are proved much better than conventional forecasting methods 本文从宏观角度和基于区域交通流小样本数据的实际情况,提出了改进的模糊指数平滑预测和中心逼近式灰色预测方法,建立了基于takagi - sugeno模糊逻辑推理的局部多元回归模型( lmrf模型) ,并进行了实例预测仿真,实例仿真结果表明改进的预测方法比传统的预测方法精度提高了好多倍。
Using the order - up - to ( out ) method and two demand forecasting patterns , that are exponential smoothing forecasting and moving average forecasting , we give the frequency response plot and the noise bandwidth figure with the help of the system control tool of matlab . we show that information sharing helps to reduce the bullwhip effect , especially at higher levels in the chain . however , the bullwhip effect problem is not completely eliminated and it still increases as one moves up the chain 本文的重点内容就是用控制论的理论和方法来研究牛鞭效应,应用补充到目标库存策略( out )和两种不同的需求预测方法,即指数平衡预测法和移动平均数预测法,用matlab的系统控制工具箱作为分析工具,分别给出了信息共享情况下和无信息共享情况下的频率响应图和带宽比较图,证明了信息共享能够减弱牛鞭效应,尤其是在供应链的高级阶段,但牛鞭效应并不能完全消除,仍旧随着供应链阶段的上升而增加。
This paper introduces two major prediction techniques , i . e . triple exponential smoothing prediction technique and gray prediction technique , and explains each technique ' s algorism and model construction in detail . by the combination of the above two method , the writer makes up the forecasting model of container throughput in shanghai port and forecasts the container throughput of near future . this would be the key content of the paper 文中详细介绍了两种预测方法各自的特点、算法及模型结构,并对上海港港口集装箱吞吐量进行了实际预测,将预测的结果与实际吞吐量作比较,从而分析两种方法的优缺点,然后将这两种单项预测法整合成一种基于有效度的组合预测法,这也是本文的核心内容。
Exponential smoothing is a technique that can be applied to time series data, either to produce smoothed data for presentation, or to make forecasts. The time series data themselves are a sequence of observations.