考慮產品生命週期末端之服務性零組件最後訂購模型 - 以汽車產業為例

Abstract

隨著產品市場競爭日趨激烈，服務性零組件重要性日益提高，它不僅是產品服務的一小部分，更是未來企業間競爭的重要因素之一。由於零組件服務期間往往比產品生產期間還長，而在產品停止生產後，零組件所需持續服務的期間通常佔整個服務期間的一大部分，特將此期間稱為末端服務期間。由於生產批量的經濟考量因素，零組件供應商在末端服務期間並不會持續生產至服務期間結束。因此，廠商必頇在零組件供應商停止生產前，下最後一次訂單以滿足未來需求，若服務廠商訂購太多將產生滯銷存貨，訂購太少則會有顧客抱怨與銷貨損失。 因此，本研究希望建立一套更有效的需求預測模式，結合報童模型極小化呆料與缺貨成本，使在面臨最後訂單問題的服務廠商能以更具成本效益的訂貨方式進行訂貨決策。首先，本研究建立一位於生命週期衰退期的零組件需求預測模式，並在本模式中考量車型流通量與回廠率，以預估零組件需求量。接著，建立剩餘年限的需求分配，並估計呆料與缺貨成本建立報童模型，透過極小化呆料成本與缺貨成本求得最適最後訂單量。最後，本研究以個案公司的實際零組件需求資料作分析研究，並以模擬驗證的方式比較本研究所建立的最後訂單方法與個案公司現行方法之優劣。 從模擬結果顯示，個案公司現行方法會傾向訂購過多零組件，使呆料成本高昂，而本研究建構的方法缺貨量與滯銷存貨皆低，總成本亦遠低於現行方法，顯示本研究所建構之方法的效果是顯著較佳的。

The importance of service parts is much emphasized as the competition in the market becomes severe. These service parts are not only providing product after-sale service but also playing one of the important rules as competing with competitors at the future. Because the duration of the service period is typically much longer than the production period, the service period after the end of production would occupy the larger proportion of the entire service period. This period is called the end-of-life service period. Especially, the supplier of service parts may no longer manufacture the parts after the certain period of time before the end of service period for considering economic production quantity. Therefore, the after service providers must place a final order to meet future demand before the production of service parts has been stopped. If the final order is larger than actual future demand, the surplus would become obsolete inventory; otherwise, it would cause customer complaint, and incur lost sales. In this paper, we first combine an effective demand forecasting model by tradeoff obsolete cost and lost sale cost considering product sales and the rate of return with newsboy problem to determine the optimal final order quantity. Next, we use the real data providing by T company to construct a simulation model in order to comparing our model with current practices of T company. The result shows our model reduce the total cost significantly, so we suggest that T company can apply the new model we developed to place its final order and control its inventory policy more effective.