臺灣博碩士論文加值系統

近兩年來國際間受到美中貿易戰以及COVID-19肺炎疫情之影響，全球經濟面臨極大的考驗，各國因疫情持續升溫，陸續將防疫等級提升，因而造成經濟活動停擺、需求趨緩，產業的生產動能也受到嚴重影響，全球各地區的製造業景氣同步衰退，企業如想在這嚴峻的環境下繼續生存，勢必要將生產成本降低、做出更有效率的管理，提升企業的競爭力，方能取得生存的機會。
然而，任何一個企業，為了滿足顧客與多變的市場需求，不管是成品或原物料都必須備有一定的存量。因此，存貨管理對整個企業發展是非常得重要。本研究個案N公司，正是面臨如何在管控居高不下的存貨成本，而又必須滿足市場與顧客及時的訂單需求，兩者之中無法取得平衡點。相信這也是大部分企業所面臨的痛點。
所以，需求預測的準確與否就扮演著相當重要的角色，若需求預測誤差過大，將無法對存貨成本及生產效能作有效的規劃管理；反之，則可能造成服務水準下降無法如期交貨，影響企業的商譽及信任度。本研究運用科學的方法（R語言工具、時間序列預測分析法）對銷售及生產時程進行預測。透過大數據資料分析並建立預測模型，對未來生產時序與採購規劃提出預測。協助企業經理人以「需求導向（Demand-Driven）」做為決策依據，進而，達到更佳的成本控管。

In the past two years, internationally affected by the U.S.-China trade war and the COVID-19 pneumonia epidemic, the global economy has faced great challenges. As the epidemic continues to heat up, countries have successively upgraded their epidemic prevention levels, resulting in the suspension of economic activities, slowing down of demand, and industrial growth. Production momentum has also been severely affected. The manufacturing boom in all regions of the world has simultaneously declined. If companies want to survive in this severe environment, they must reduce production costs, make more efficient management, and enhance their competitiveness. Can get a chance of survival.
However, in order to meet the needs of customers and the ever-changing market, any company must have a certain stock of finished products or raw materials. Therefore, inventory management is very important to the development of the entire enterprise. In the case of this study, Company N is faced with how to manage the high inventory cost, while also having to meet the market and customers' timely order demand, which cannot strike a balance between the two. I believe this is also a pain point faced by most companies.
Therefore, the accuracy of the demand forecast plays a very important role. If the demand forecast error is too large, it will not be able to effectively plan and manage inventory costs and production efficiency; on the contrary, it may cause the service level to drop and fail to deliver on time. The goodwill and trust of the company. This research uses scientific methods (R language tools, time series forecasting analysis method) to forecast sales and production schedules. Through the analysis of big data and the establishment of forecasting models, forecasts are made for future production timing and procurement planning. Assist corporate managers to make "Demand-Driven" decisions to achieve better cost control.

謝誌·································································Ⅰ
論文摘要·····························································Ⅱ
Abstract·····························································Ⅲ
目錄·································································Ⅴ
表目錄·······························································Ⅶ
圖目錄·······························································Ⅷ
第一章 緒論·························································1
第一節 研究背景與動機················································1
第二節 研究目的·····················································2
第三節 研究流程·····················································3
第二章 文獻探討·····················································5
第一節 時間序列預測法················································5
第二節 庫存管理·····················································11
第三章 研究架構與方法················································15
第一節 研究個案公司簡介··············································15
第二節 變頻器產業概況與產業趨勢分析···································18
第三節 研究設計·····················································25
第四節 研究方法與資料分析············································26
第四章 研究結果·····················································47
第一節 預測準確率分析················································47
第二節 研究個案N公司訂單趨勢與時間序列類型分析························50
第三節 研究個案N公司2021年訂單預測分析·······························53
第五章 結論與建議···················································55
第一節 結論························································55
第二節 建議························································56
參考文獻····························································58

一、中文文獻
1.方法茲(2020)。使用時間序列分析和R中的機器學習預測“銀行部門”BBRI和BBNI的股價。正修科技大學工業工程與管理研究所，碩士論文，未出版，高雄市。
2.王睦華(2018)。應用時間序列法及資料探勘技術於精實生產管理之研究-以B公司之產品銷售預測為例。國立虎尾科技大學工業管理系，碩士論文，未出版，雲林縣。
3.王威勝、陳詠祥、曹志誠 、林冠誠、陳偉霖(2006)。交換式電源供應器。樹德科技大電腦與通訊系，學士論文，未出版，高雄市。
4.石育賢(1999)。全球變頻器市場趨勢與經營策略分析。工研院IEK系統能源組。
5.李琴容（2010）。淺談產銷管理。運籌總處/產銷管理部。企管新語-連展人。5月號。
6.林昆平(2005)。我國交換式電源供應器市場現況及發展趨勢。經濟部標準檢驗局。
7.林茂文(1992)。時間數列分析與預測。華泰書局。
8.林家慶(2009)。小型製造業訂單需求預測方法研究─以台中C公司生產現場為例。亞洲大學資訊科學與應用學系，碩士論文，未出版，台中市。
9.林慶仁(2001)。最新的電力電子轉換技術與產品特點。新電子科技雜誌第188 期2001‧11 月號。
10.秦志遠(2013)。挑戰領導品牌變頻器之競爭策略-以T公司為例。元智大學管理碩士在職專班，碩士論文，未出版，桃園市。
11.秦劍雲（1989）。知識庫支持之訂貨生產排程系統。國立台灣科技大學工業管理研究所，碩士論文，未出版，台北市。
12.張政雄(2010)。銷售量應用於訂購預測模式之研究。大同大學工程管理碩士在職專班，碩士論文，未出版，台北市。
13.溫子儀（2018）。製造業如何做好最適庫存管理之推動。管理知識中心。
14.楊振隆、蔡志弘、黃楣棋（2006）。市場導向、製造彈性對製造優勢影響關係之實證研究—以印刷電路板業為例。中華管理評論國際學報11月。
15.廖子儀(2018)。原物料商品價格之預測-以ARIMA模型分析。國立高雄應用科技大學國際企業研究所，碩士論文，未出版，高雄市。
16.鄭永賢(2006)。存貨控制模式之研究－以台電集管材料為例。中華大學經營管理研究所，碩士論文，未出版，新竹市。
17.劉子琪(2013)。以限制理論於企業生產流程系統改善─從銷售預測、生產至庫存。東海大學工業工程與經營資訊研究所，碩士論文，未出版，台中市。
18.劉代華(2008)。訂單需求預測-以時間序列分析研究。國立清華大學工業工程與工程管理學系，碩士論文，未出版，新竹市。
19.戴群諺(2014)。變頻器的原理與應用。國立宜蘭大學機械與機電工程學系，碩士論文，未出版，宜蘭縣。
20.GaryHsia(2016)。人才培訓網。<吃掉企業獲利的兇手！產品經理該如何做好「庫存管理」>。https://www.smartlinkin.com.tw/article/2751。
21.Pamela Huang(2019)。SAS Taiwan。<讓 AI 入魂。精準預測你的需求與庫存>。https://www.sas.com/zh_tw/insights/analytics/embedded_ai_predict_demand_and_inventory_precisely.html
二、英文文獻
1. Alameer, Z., Fathalla, A., Li, K., Ye, H., & Jianhua, Z. (2020). Multistep-ahead
forecasting of coal prices using a hybrid deep learning model. Resources Policy,
65,101588.
2. Bildirici, M. E., & Sonustun, F. O. (2018). The effects of oil and gold prices on
oil-exporting countries. Energy Strategy Reviews, 22, 290-302.
3. Cooper, M. C. and Lisa, M. E.( 1993). Characteristics of Supply Chain Management and
Implication for Purchasing and Logistics Strategy. The International Journal of
Logistics Management.
4. Delurgio, Stephen A. (1999). Forecasting Principles and Applications, 1st edn, Business
Forecasting, 5th edn. McGraw-Hill.
5. Deanna Diehn(2001) . Seven Steps to Build a Successful Collaborative Forecasting
Process. 19(4).
6. Frigstad, David (1996). Industrial Market Research & Forecasting. Frost & Sullivan.
7. Guoqiang Peter Zhang(2001). An investigation of neural networks for linear time-series
forecasting, Computers and Operations Research, Vol. 28.
8. Hamilton , James D. (1994). Time Series Analysis. Princeton University Press.
9. Jinquan Guo, Hongwen He and Chao Sun. (2019). ARIMA-Based Road Gradient and
Vehicle Velocity Prediction for Hybrid Electric Vehicle Energy Management. IEEE
Transactions on Vehicular Technology, pp.5309-5320, 2019.
10. Kelley, S. W.( 1992). Developing Customer Orientation among Service Employees.
Journal of the Academy of Marketing Science, 20, No.1.
11. Kalid Yunus, Torbjörn Thiringer, and Peiyuan Chen. (2016). ARIMA-Based
Frequency-Decomposed Modeling of Wind Speed Time Series. IEEE Transactions on
Power Systems, pp.2546-2556.
12. Lewis, C. D.(1982). Industrial and Business Forecasting Method. London:
Butterworth-Heinemann.
13. Mohammad Almasarweh & S. AL Wadi.(2018). ARIMA Model in Predicting Banking
Stock Market Data, Canadian Center of Science and Education Vol. 12.
14. Masum, S., Liu, Y., & Chiverton, J. (2018). Multi-step time series forecasting of electric
load using machine learning models. In International Conference on Artificial
Intelligence and Soft Computing (pp. 148-159). Springer, Cham.
15. Wilson, J H and Keating, B (2007). Business Forecasting, 5th edn. McGraw-Hil.