近年來台灣汽車市場已漸漸達致飽和,國內外各汽車業者無不以新車款及特價等方式吸引消費者,然而,台灣的汽車市場在規模上就有先天的限制,因此,汽車業者皆將其目光轉移到汽車售後服務市場,因此在汽車市場之外,售後服務市場也漸漸成為汽車業的新寵,除了原廠車的零件廠,還有其他各種副廠零件紛紛出爐,競相搶食汽車零件售後服務這塊大餅。 本研究將目標放在汽車原廠零件的庫存管理。藉由實務上的例子,由零件需求面的角度進行分析,由於汽車零件種類及數量繁多,而一方面要有充份的庫存以隨時因應不時之需,另一方面又要能盡量壓低庫存,增加存貨週轉,如何管理如此龐大繁雜的零件數目,便成了售後服務零件廠主要的課題之一。一般在做庫存管理之前必須先分類,將各零件歸類畫分,再依據各零件的分類做零件的控管。本研究參考現行的庫存管理辦法,提出一套新的庫存管理模式,先按需求變動型態做為分類依據,另外再提出利用這個分類方式產生不同零件安全庫存乘數,並結合新的月平均需求量估算方式以得到較佳的庫存上限。 最後比較現行與本研究所提出的建議這兩種方式在實際應用上的和優劣勢,藉由eM-Plant軟體模擬此兩種不同的庫存管理方式,再透過模擬所得到的結果分析這兩種方式的差異。透過系統模擬的分析,可以瞭解按需求變動分類的合理性,幫助上游零件服務廠瞭解如何運用有邏輯系統的分類方式歸類零件,透過模擬的結果掌握零件需求變動型態。
In recent years, the vehicle market is saturated step by step and the automobile factories whether domestic or foreign attracted consumers by selling new-designed cars and having a special offer. However, the scale of Taiwan vehicle market is limited. The automobile factories have already had their attention on the after-market, and the after-market is more and more important. Other than the OEM plant, there are many other part plants devoting into this market. We focused on the stock management of OEM plants. By a practical case, we analyzed the inventory management from demand aspect. Because of the large quantities and numerous items, we must have enough stocks in order to meet the accidental demands; on the other hand, we must decrease stocks and increase the turnover rate of inventory. So, the main subject we focus on is how to manage the large number of service parts. We proposed a method to cluster the part items by demand variation in accordance with the practical method, and constructed a new method to calculate the safety stock for demand and monthly average demand. Finally, we can calculate the better stock limits. Finally, we compare the pros and cons between out proposed and current practice, and then we simulated by eM-Plant software. We can analyze the two outcomes to know what differences between two methods. Through the simulation, we can know about the rationality of clustering by demand variation, and we can help the part plants know how to use a logical and systematic clustering method to cluster the items, and we can know about the variation pattern of the part demands.