國際來台旅遊需求預測模型之先期研究

A Preliminary Study of Forecast Models for International Taiwan Inbound Tourism

陳泉蒲(Chuan-Pu Chen)

指導教授 ： 王銘宗

國立臺灣大學/工學院/工業工程學研究所/碩士(2014年)

https://doi.org/10.6342/NTU.2014.10898

摘要

隨著台灣經濟的快速增長，旅遊業已成為國家發展的一個重要的新領域，對國家經濟的發展進行全面性的刺激作用。觀光人口屢屢突破每年人次高峰，隨著此趨勢觀光產業也迎向新的發展階段，觀光旅遊業只要成功吸引外國遊客使旅遊服務的需求增加，能夠為國家的經濟發展、貿易表現作提供重大貢獻。因此，瞭解何種變數會影響旅遊需求以及如何應用這些變數預測國際觀光旅客的來台需求趨勢和預測，給予決策者有效資訊來做策略規劃是非常重要的。
人工智慧技術已經成為經濟建模和預測的重要工具，隨著資訊科技的發展、預測技術的演進與資料的更新，多項新技術已經應用在旅遊需求預測。然而，就預測準確性而言，研究分析顯示始終沒有一個最佳的模型可在多個情形下統一適用和絕對優於其他模型。本研究重新針對預測模式的精確性與適用性進行探討，主要以倒傳遞類神經網路以及支撐向量迴歸模型於國際旅遊需求預測。
本研究主要的貢獻在於透過實證性研究，期望藉由本研究分析結果，對於觀光來台人數預測，有進一步的了解，促進觀光產業對預測模型之了解與共識，並給予觀光產業策略建議，作為觀光領域或預測模型後續研究發展之參考。

關鍵字

旅遊需求預測 ； 類神經網路 ； 倒傳遞神經網路 ； 支撐向量回歸

並列摘要

With the rapid growth of Taiwan's economy, tourism has become an important new areas of national development and comprehensive stimulate national economic development. Tourism population number has exceeded the annual peak year by year. With this trend, the tourism industry also forth to meet a new development phase. As long as attracting foreign tourists, increasing demand for tourism services provide a significant contribution for national economic development and trade performance. Therefore, to understand what variables will affect tourism demand and how to apply variables to predict international tourist arrivals to Taiwan, giving decision-makers information to make strategic planning is very important.
Artificial intelligence technology has become an important tool for economic modeling and forecasting. With the development of information technology, several new technologies have been applied in travel demand forecasting. However, not an optimal model can be applied uniformly and absolutely superior to other models in multiple scenarios. This research discussed the accuracy and applicability of predictive models in back propagation network and support vector regression model for international tourism demand forecasting.
It is hoped that this research could provide a better understanding about tourism demand forecasting and give the tourism industry policy recommendations.

並列關鍵字

Tourism Demand Forecast ； Artificial Neural Network ； Back Propagation Network ； Support Vector Regression

參考文獻

Aaker, D. A. (1998). Strategic market management. John Wiley & Sons.

Aburto, L., & Weber, R. (2007). Improved supply chain management based on hybrid demand forecasts. Applied Soft Computing, 7, 136–144.

Athiyaman, A., & Robertson, R. W. (1992). Time series forecasting techniques: short-term planning in tourism. International Journal of Contemporary Hospitality Management, 4.

Ayeh, J. K., & Lin, S. (2011). “Estimating tomorrow”s tourist arrivals’: forecasting the demand for China’s tourism using the general-to-specific approach. Tourism & Hospitality Research, 11, 197–206.

Box, G. E. P., & Jenkins, G. M. (1970). Time Series Analysis Forecasting and Control,.