Demand Forecasting with Five Parameter Exponential Smoothing

11th Curtin University Technology, Science and Engineering (CUTSE) International Conference

工业技术（总论）

Ratnasari, S.^1 ; Yuniaristanto^2 ; Zakaria, R.^2

Logistics System and Business Laboratory, Universitas Sebelas Maret Surakarta, Indonesia^1

Department of Industrial Engineering, Universitas Sebelas Maret Surakarta, Indonesia^2

关键词: Daily time series;    Demand forecasting;    Economic development;    Exponential smoothing;    Exponential smoothing method;    Forecasting methods;    Holt-Winters method;    Newspaper industry;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/495/1/012014/pdf
DOI  :  10.1088/1757-899X/495/1/012014

学科分类：工业工程学

来源: IOP

【 摘 要 】

Changes in technology has affected many aspects on economic development. One of them is Newspaper Industry. Reports indicate that globally the newspaper is passing through its hardest time ever. Therefore, newspaper industry need to be more creative in order to be able to deal with change and maintain its existence. In supply chain management newspaper industry, forecasting is the most important part to predict the future demand, minimize waste of product, scheduling production, optimize inventory level and resources. The exponential smoothing methods are simple but the best approach and popular methods used for forecasting. This study aimed to implement five parameter exponential smoothing to predict number of newspaper demand in the future with various method to get the best forecast result. The perfomance of method is evaluated using a newspaper demand daily time series. MSD and MSE used to determining and selecting the best forecasting method. The result show that additive Holt winters method with damped trend (DAHWM) is suitable for demand forecasting in newspaper industry.