- 學位論文
汽車零組件存貨預測模式之研究-以W公司為例
A Study of the Automobile Parts Inventory Forecast Model - A Case of W Company
摘要
近年來，汽車設計更加考慮人性與使用性，帶動了車用電子的產品快速成長。預估2020年全球車用電子的產值將可達到3,830億美元，對台灣汽車零組件廠商是項佳音。汽車零組件產業想在國際市場上競爭，除了技術能力外，少量多樣生產、減少存貨、提升周轉、降低成本及增進現金流量，都是提升競爭力的途徑。 本研究以某車用電子零組件協力廠商為例，探討該個案公司的存貨管理政策。使用個案公司2014年及2015 年第一季的銷售資料，作為預測分析之基礎。所使用的預測模型為移動平均法、加權移動平均法、指數平滑法及迴歸分析等，以均方誤差(MSE)最為比較預測優劣之指標。 驗證結果顯示迴歸分析的均方誤差最低，其次為加權移動平均法、移動平均法，而指數平滑法的均方誤差最高。研究結果較個案公司經驗值誤差低。系統方法與客觀數據，值得推薦給個案公司作為存貨管理上的範本。
並列摘要
The markets of electronic automotive products are growing fast because the design of the automotive intend for convenience and user friendly in recent years. The estimation of market value of worldwide automotive parts will reach to USD 383 billion dollars in 2020. It is a positive driven force in for automotive parts manufacturers in Taiwan. In order to have strong competition ability in global market, beside the technical capabilities, production approaches of small volume, various styles, minimization of stock, higher inventory turnover rate, cost reduction, and smooth cash flows are all necessary. This study uses an automotive parts supplier as subject and studies her policy of inventory management. The data used in the prediction analysis is based on the sales information in 2014 and Q1 2015 of the subject company. The prediction methods used include the Moving Average Method, the Weighted Moving Average Method, the Exponential Smoothing Method and the Regression Analysis. Mean Squared Error (MSE) is used as indicator to compare the analytical results. The results showed that the Regression Analysis has lowest MSE followed by the Moving Average Method and the Weighted Moving Average Method. The Exponential Smoothing Method has the largest MSE. All these errors (MSE), however, are lower than the values currently used in this case company. The systematic analysis and objective results are useful for the case company as templates for inventory management.