銷售量應用於訂購預測模式之研究

ORDERING FORECASTING MODEL BASED ON PREVIOUS SALE DATA

張政雄(Cheng-Hsiung Chang)

指導教授 ： 黃然 ； 翁在龍

大同大學/工程學院/工程管理碩士在職專班/碩士(2011年)

摘要

預測銷售量是企業永續經營的重要課題。本研究的目的是提出一個適合的銷售預測方法和訂購管理，進而提早交期、降低成本及減少庫存。
本研究利用移動平均法、指數平滑法、迴歸分析法及時間序列分析法作為分析工具，進行銷售量預測。利用個案A公司電源模組產品季銷售的數據作計算，將結果的預測值與實際銷售量比較，利用平均絕對誤差(MAE)、誤差均方根(RMSE)及平均絕對誤差百分比(MAPE)來判斷預測值的準確度，並以此預測值作為訂購及存貨管理的依據。
根據比較的結果，指數平滑法為誤差值最低，而時間序列分析法為誤差值最高。

關鍵字

移動平均法 ； 指數平滑法 ； 迴歸分析法 ； 時間序列分析法

並列摘要

Sale prediction is an important issue for a firm to survive in the competition market. The aim of this study is to propose a suitable predictable method for sale and ordering management for accelerating the delivery of products and reducing the cost by minimizing the inventory. .
Moving average method, exponential smoothing method, regression analysis and time series analysis method were used as analytical tools for sale prediction. Case study was based on the data of an electric company A which produced various types of converters. The computed results were compared with the sales data. Mean average error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE) were selected to evaluate the error for prediction accuracy.
Based on the comparative results, exponential smoothing method has the lowest error and time series analysis method has the highest error.

並列關鍵字

moving average ； exponential smoothing ； regression analysis ； time

參考文獻

﹝14﹞林茂文，「時間數列分析與預測」，華泰文化事業股份有限公司，2006年。

﹝15﹞吳柏林，「時間數列分析導論」，華泰文化事業股份有限公司，1995年。

﹝6﹞Donlebell and O.J.Krasner, ”Selecting Environmental Forecasting from Business Planning Requirements,”Academy of Management Review, 373-383, July 1997.

﹝9﹞William J. Stevenson,”Operatins management.8th ed,” NewYork：The McGraw-Hill companies, 2005.

﹝10﹞Simchi-Levi, D.,Kaminsky, P., and Simchi-Levi, E , ”Designing and Managing the Supply Chain:Concepts, Strategies, and Case Studies, ” McGraw-Hill: Irwin.