臺灣博碩士論文加值系統

近年來全球航運中心有東移現象，臺灣四面環海，海運成為連結對外貿易發展之命脈，更為國內經濟主要核心競爭力之一。本文主要以德國航運經濟與物流研究中心(ISL) 2002~2017年臺灣與新南向三國(新加坡、印度與印尼)船隊資料為主、以聯合國UNCTAD (2002~2017)資料為輔。運用基本統計分析，比較臺灣與新南向三國船隊結構之變化，並透過灰預測GM(1,1)與循還式殘差修正(Recursive residual)提高精確度，分析臺灣與新南向三國整體船噸發展趨勢，並藉平均絕對誤差(MAE)、平均絕對誤差百分比(MAPE)與殘差均方根(RMSE)進行精確度比較。本文主要研究發現如下：

(一) 就總體船噸而言，近年臺灣與新南向三國不論船舶艘數或載重噸均呈現逐年遞增趨勢，由船噸大小可知，確實有船舶大型化之趨勢，其中以臺灣最為明顯。於船舶設籍方面，印尼因有17,000多個島嶼及沿海航行權因素所致，岀籍比率僅12%，新加坡、臺灣、印度與全球出籍比率均逐年攀升，其中以臺灣最為顯著超過91%。於船齡方面，新加坡、印度與臺灣船齡與全球一致，確實有年輕化之趨勢，印尼船舶最為老舊平均船齡21.4年。於船型與載重噸方面，全球目前以油輪與散裝船為大宗，新加坡、印度與印尼主要擁有油輪，臺灣主要擁有散裝船與貨櫃船。就平均船價而言印尼船舶艘數最多但平均船齡較高，以致平均每艘船價值相對最低，而臺灣總載重噸相對較多，或因經營散裝船為主，以致於平均每載重噸價值最低。

(二)有關臺灣和新南向三國整體船噸預測，灰預測結果顯示，RRGM(1,1)模型與GM(1,1)模型相較均適用於預測船噸發展。於預期效能方面，經殘差檢驗後，不論是否經殘差修正，其平均精確度均高於90%，顯示灰預測GM(1,1)與RRGM(1,1)模型，對臺灣與新南向三國整體船噸發展之預測，均為效能良好之預測模型。若全球經濟無重大震盪變化前提下，就未來整體船噸成長預測而言，至2020年時，新加坡降至54,421千載重噸，臺灣增為52,594千載重噸，印度增為23,390千載重噸，印尼增為20,496千載重噸，綜言之，臺灣、印度、印尼船噸均成長，而新加坡呈現負成長之發展趨勢。

上述研究結果，提供相關航運業者於未來經營管理策略規劃之依據，以及協助政府主管機關於以擬定相關航運政策之參考。

In recently years, the global shipping market has moved to Eastern. Since Taiwan is an island surrounded by the sea, shipping is not only the lifeblood for the development of foreign trade but also the competitive edge for the domestic economy. The article is mainly based on the data from the Institute of Shipping Economics and Logistics (ISL) from 2002 to 2017 of the statistics from the New southbound Shipping Countries and Taiwan fleet, supplemented by UNCTAD (2002~2017). Using basic statistical analysis to compare changes in the structure of the New southbound Shipping Countries and Taiwan fleet, and to improve the accuracy of the GM (1,1) and the recursive residuals through grey forecasting. Then, analyze the overall development trend of the tonnage between Taiwan and the New South three countries. After that, use the mean absolute error (MAE), mean absolute error percentage (MAPE), and root mean squared error (RMSE) for accuracy comparison. The main research founded in this paper are as follows:

1.In recent years, the overall tonnage both Taiwan and New South three countries have been present year by year regardless of the number of vessels or deadweight tonnage. According to the increase in the size of the ship, there is indeed a trend of increasing ship sizes in all of them, but Taiwan is the most obvious. In Indonesia, there are more than 17,000 islands and coastal navigation rights. Though the current registration ratio is only 12%, to compare the rising rates of registration with other Asia countries such as India and global rates, Taiwan is the most significant one. In terms of age, to compare the ship ages of Singapore, India and Taiwan Consistent with the world, another trend of having younger vessels, Indonesian ships are the oldest. In terms of ship type and deadweight tonnage, the world is currently dominated by tankers and bulk carriers. Singapore, India and Indonesia mainly own oil tankers, and Taiwan mainly owns bulk carriers and container ships. If we examine the average shipping price, Indonesia has the largest number of vessels, but the average age is relatively high. Consequently, the average value of each vessel is relatively the lowest, while the total load in Taiwan is relatively high. It is inferred that the bulk carriers are mainly operated, resulting in average load per vessel and the lowest value of tonnage.
 
2.For the prediction of overall ship tonnage in Taiwan and the New South three countries, the grey forecast results show that both the RRGM (1,1) model and the GM(1,1) model are suitable for predicting ship tonnage development. In terms of expected performance, after testing the residuals, whether differential correction, with an average accuracy of over 90%, shows the grey prediction GM (1,1) and RRGM(1,1)The model predicts the overall development of ship tonnage in Taiwan and the New South three countries. If there is no major shock in the global economy, in terms of the overall growth forecast for ship tonnage, by 2020, Singapore will drop to 54,421 thousand deadweight tons, Taiwan will increase to 52,594 thousand deadweight tons, and India will increase to 23,390 thousand dwt, Indonesia. Increased to 20,496 thousand dwt. In summary, the ship tonnage in Taiwan, India, and Indonesia grew, with the only Singapore showing negative growth.

The above findings provide the basis for the relevant shipping industry's future business management strategy planning, and assist the government's competent authorities in formulating relevant shipping policies.

謝誌 I
摘要 II
ABSTRACT III
目錄 V
圖目錄 VII
表目錄 VIII
第一章 緒論 1
1.1 研究背景與動機 1
1.2 研究問題與目的 3
1.3 研究內容與方法 4
1.4 研究架構與流程 5
第二章 文獻回顧與評析 8
2.1 航運政策與船舶設籍因素 8
2.2 船噸結構相關文獻 9
2.3 灰理論應用於航運領域相關文獻 13
2.4 綜合評析 13
第三章 臺灣與新南向三國船隊現況分析 15
3.1 船舶艘數與載重噸分析 15
3.2 國輪與外輪和船齡分析 20
3.2.1 國輪與外輪艘數與載重噸 20
3.3.2 船齡分析 23
3.3 船舶設籍分析 24
3.4 主要船型與載重噸分布 25
3.5 船舶艘數與載重噸和船價分析 27
3.6 本章小結 28
第四章 研究方法 29
4.1 灰色理論 29
4.1.1 級比檢定 29
4.1.2 灰預測GM(1,1)模型 30
4.1.3 灰預測RRGM(1,1)模型 32
4.1.4 誤差分析 33
4.2 精確度衡量 35
4.2.1 平均絕對誤差 35
4.2.2 平均絕對誤差百分比 36
4.2.3 殘差均方根 36
第五章 整體船噸預測實證分析 37
5.1 灰色理論應用 37
5.1.1 級比檢定 37
5.1.2 灰預測精確度分析 38
5.3 灰預測結果比較 39
5.4 預測精確度比較 40
5.5 區間預測結果分析 41
5.6 綜合討論 42
第六章 結論與建議 43
6.1 結論 43
6.2 建議 44
參考文獻 46
 

1. 包嘉源(2014)，「亞洲鄰近國家海運發展政策分析」，航運季刊，第二十三卷第一期，頁51-67。
2. 朱子昕(2009)，散裝船舶大型化及市場總船噸之影響因素與未來趨勢分析，國立交通大學運輸與物流管理學系所碩士論文。
3. 李佳逸、張志清、包嘉源、呂頌揚(2009)，「由各國航運稅制論我國因應之道」，航運季刊，第十八卷第三期，頁91-111。
4. 許亢亭(2012)，運用灰預測模型探討松山機場兩岸直航客運量發展，中華科技大學航空運輸研究所碩士論文。
5. 林繼國、陳一平(2002)，國輪船隊整建問題之研究，交通部運輸研究所，頁1-84。
6. 吳偉銘、陳奕誠(2009)，「二手船舶價格指數之建構—Hedonic價格迴歸模型之應用」，運輸學刊，第二十一卷第一期，頁1-23。
7. 許巧鶯、溫裕弘(1997)，「臺灣地區國際航空客運量之預測—灰色預測模式之應用」，運輸計劃季刊，第二十六卷第三期，頁525-556。
8. 馬豐源(2005)，「以灰色系統理論探討海運成長率預測與經濟成長率關聯性」，石油季刊，第四十一卷第四期，頁71-80。
9. 馮正民、鍾政棋、袁劍雲(2002)，「船舶設籍對航商船員成本之影響」，運輸計畫季刊，第三十一卷第三期，頁663-678。
10. 羅仕京(2014)，影響大眾運輸系統運量之社會經濟因素灰預測模式之研究，中華大學運輸科技與物流管理研究所碩士論文。
11. 黃承傳、鍾政棋(2005)，「我國散裝船舶設籍關鍵影響因素之分析」，運輸計劃季刊，第三十四卷第四期，頁426-450。
12. 傅立(1992)，灰色系統理論及其應用，第一版，北京：科學技術文獻出版社。
13. 鄧聚龍(2000)，灰色系統理論與應用，臺北市：高立圖書公司。
14. 鄧聚龍、郭洪(1996)，灰預測原理與應用，臺北市：全華科技圖書(股)公司。
15. 鍾政棋、徐嘉陽(2007)，「兩岸三地整體船噸結構及其成長趨勢之分析」，運輸計劃季刊，第三十六卷第一期，頁426-450。
16. 鍾政棋、徐穎珍(2011)，「灰色理論應用於兩岸三地貨櫃船運能供給之預測」，航運季刊，第二十一卷第一期，頁47-70。
17. Alderton, T. and Winchester, N. (2002), “Flag States and Safety: 1997-1999,” Maritime Policy and Management, Vol. 29, Iss. 2, pp. 151-162.
18. Bergantino, A.S. and O’Sullivan, P. (1999), “Flagging Out and International Registries: Main Developments and Policy Issues,” Journal of International Transport Economics, Vol. 26, No. 3, pp. 447-472.
19. Cullinane, K. and Robertshaw, M. (1996), “The Influence of Qualitative Factors in Isle of Man Ship Registration Decisions, Vol. 23, Iss. 4, pp. 321-336.
20. Cheng, P.L., Huang, W.C., Yu, Z.H., and Wu, S.C. (2003), “Application of Grey Theory to the Container Cargo Transport Demand Forecast in Port,” Proceedings of the 25th Ocean Engineering Conference in Taiwan, October, National Taiwan Ocean University, pp. 849-855.
21. Chung, C.C., Hsu, C.Y., and Hwang, C.C. (2009), “The Vessel Tonnage Supply and Developmental Trends of China, Hong Kong, Singapore and Taiwan,” The Journal of Grey System, Vol. 21, Iss. 2, pp. 137-148.
22. French, C.J. (1973), “Eighteenth Century Shipping Measurements,” The Journal of Economic History, Vol. 33, No. 2, pp. 434-443.
23. Haralambides, H.E. and Yang, J. (2003), “A Fuzzy Set Theory Approach to Flagging Out: Toward a New Chinese Shipping Policy,” Marine Policy, Vol. 27, Iss. 1, pp.13-22.
24. Hsu, L.C. (2003), “Applying the Grey Prediction Model to the Global Integrated Circuit Industry,” Technological Forecasting and Social Change, Vol. 70, Iss. 6, pp. 563-574.
25. Hoffmann, J., Sanches, R.J., and Talley, W.K. (2005), “Determinants of Vessel Flag,” Research in Transportation Economics, Vol. 12, Iss. 1, pp. 173-219.
26. ISL (2002-2017), Institute of Shipping Economics and Logistics, Shipping Statistics and Market Review, ISL/SSMR, Germany.
27. Jung, B.M. and Kim, S.J. (2012), “Change of Shipping Industry Circumstances and Shipping Policy Directions of Developing and Developed Countries,” The Asian Journal of Shipping and Logistics, Vol. 28, Iss. 2, pp. 135-160.
28. Lewis, C.D. (1982), International and Business Forecasting Method, London: Butterworths Scientific.
29. Lin, C.T., Wang, S.M., and Chiang, C.T. (2001), “Manpower Supply and Demand of Ocean Deck Officers in Taiwan,” Maritime Policy and Management, Vol. 28, Iss. 1, pp. 91-102.
30. Llácer, M., Francisco, J. (2003), “Open Register: Past, Present and Future,” Marine Policy, Vol. 27, Iss. 6, pp. 513-523.
31. Merra, S.N. (2005), “Development of Indian Shipping during Five Year Plans,” Finance India, Vol. 19, No. 4, pp. 1337-1364.
32. Marlow, P. and Mitroussi, K. (2008), “EU Shipping Taxation: The Comparative Position of Greek Shipping, “Maritime Economics and Logistics, Vol. 10, Iss. 1-2, pp. 185-207.
33. Pi, D., Liu, J., and Qin, X. (2010), “A Grey Prediction Approach to Forecasting,” Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, Vol. 32, Iss. 16, pp. 1517-1528.
34. Piniella, F., Alcaide, J.L., and Rodriguez-Diaz, E. (2017), “The Panama Ship Registry: 1917-2017,” Marine Policy, Vol. 77, March, pp. 13-22.
35. UNCTAD (2002-2017), Review of Maritime Transport, United Nations Conference on Trade and Development, United Nations, UNCTAD/RMT, New York and Geneva.
36. Yin, J., Fan, L., and Li, K.X. (2018), “Second Ship Registry in Flag Choice Mechanism: The Implications for China in Promoting a Maritime Cluster Policy,” Transportation Research Part A, Vol. 107, January, pp. 152-165.
37. Yin, J. and Fan, L. (2018), “Survival Analysis of the World Ship Demolition Market,” Transport Policy, Vol. 63, April, pp. 141-156.
網頁資料
1. 行政院(2016)，行政院新聞傳播處，網址：https://www.ey.gov.tw/Page/9277F759E41CCD91/87570745-3460-441d-a6d5-486278efbfa1，擷取日：2018年5月15日。
2. 科技政策研究與資訊中心(2017)，「以新加坡作為前進東南亞之創業基地」，網址：https://portal.stpi.narl.org.tw/index/article/10307，擷取日：2018年5月15日。
3. 經濟部(2018)，中華民國經濟部統計處，經濟統計指標，網址：file:///C:/Users/user/Desktop/0972252f-1da3-408a-9746-2cb5bd67d2b9.pdf，擷取日：2018年4月25日。
4. International Trade Centre (2006), Trade Statistics. DOI: https://www.trademap.org/tradestat/Service_SelService_TS.aspx?nvpm=1|000||||||S00|1|3|2|1|2|1|5|1|, Retrieved on May 15, 2018.