自2016年以來，在陸客限縮來台背景下，台灣觀光業面臨成長的壓力，因此政府積極推動觀光新南向政策。根據觀光局統計，2018年越南旅客來台約49萬人次，成長率高達28 %，居東南亞國家之冠;另外越南國內經濟成長穩定及人口數量龐大，具備成為來台觀光客源地之潛力及條件。為能吸引更多旅客來台，提升本國觀光服務品質成為優先議題。本研究藉由研究越南來台旅客之旅遊體驗，進而改善旅遊業者之服務品質，以增加本國觀光業競爭力。重要-表現度分析(Importance-Performance Analysis, IPA)模式廣泛運用於服務業缺失改善案例中，本文以重要-表現度分析模式改善越南來台旅客服務品質，其中劃分象限中心點方式會明顯影響屬性的象限配置，決定改善方案及企業資源分配。因此，本文選擇傳統型數據中心(Data-centered)、對角線(Diagonal line)劃分象限模式，與改良型重要-表現度缺口分析(Importance-performance and gap analysis, IPGA)、重要-表現度優先排序分析(Importance-performance and preference-rankings analysis, IPPRA)模式，評估各種模式在劃分屬性象限及排定改善次序上之差異情形。前述四種IPA模式比較結果，傳統IPA方法僅能篩選出問項所屬象限，無法排定改善次序，IPGA與IPPRA模式，可改進座標點辨識能力，有助判別所屬象限，且可依座標計算優先改善排序，此特性可彌補傳統IPA模式的限制，在實務運用上具備較佳之可用性。綜合四種模式之優先改善內容，發現有7項是共同項目，包含:「名勝古蹟與觀光景點之吸引力、觀光景點之環境整潔與衛生、台灣景點具地方及文化歷史特色、台灣提供觀光旅遊的資訊平台、提供多樣選擇之旅遊行程產品、旅行社之服務效率、來台灣旅遊的整體印象」，此結果可作為越南來台旅遊團服務品質之最優先改善內容，協助提升組織資源分配效益及策劃改善方案效能。IPGA與IPPRA模式之改善排序仍有部分差異，需更多數據去作驗證及分析。

Since 2016, due to the context of the travel restrictions on mainland China, Taiwan’s tourism industry is under pressure to grow. Therefore, the Taiwan government is actively promoting the tourism policy of New Southbound Policy. According to the statistics of the Tourism Bureau, Vietnamese tourists came to Taiwan about 490,000 passengers in 2018 with a growth rate of 28% ranking the highest in Southeast Asia countries. In addition, Vietnam’s domestic economy is growing steadily and its population is huge, thus it has the potential to become a tourist source for Taiwan. In order to attract more tourists to Taiwan, improving the quality of tourism services in the country has become a priority issue. This study improved the service quality of the tourism by studying the tourist’s experience of Vietnamese in Taiwan to increase the competitiveness of the tourism industry. Important-Performance Analysis (IPA) models are widely used in problem improvement cases of service industry. This study used IPA models to improve the service quality for Vietnamese tourist in Taiwan. The way to deside the cross-point of the quadrant may significantly affect the quadrant configuration of the attributes and prioritize the improvement, which can determine the improvement plan and allocation of enterprise resource. Therefore, this paper selected the Data-centered and the Diagonal line models to divide the quadrant. In addition, we used importance-performance and gap analysis (IPGA) and importance-performance and preference-rankings analysis (IPPRA) models to evaluate the attribute’s quadrant and improving priority, as well as to compare the difference between all the models. Compared with the above four IPA models, the traditional IPA models can only filter out the quadrant of the question, and cannot determine the order of improvement. The IPGA and IPPRA models can identify the question’s coordinate more precisely, thus can sort the quadrant and calculate the improving priority. This feature can make up for the limitations of the traditional IPA models and provide better usability in practical applications. Based on the priority improvement items of the four models, seven items were found to be the same, including: “Attraction of scenic and touring spots, cleanliness and sanitation of sightseeing spots, sightseeing attractions of local and cultural features in Taiwan, provision for sightseeing information platform in Taiwan, travel itinerary offering a wide range of choices, service efficiency of travel agencies, overall travel image in Taiwan”. These results can be used to improve the top-priority service quality of Vietnam's tourist, helping to improve the effectiveness of resource allocation and efficiency of correcting plan for travel enterprise. There are some differences in the improvement ranking between IPGA and IPPRA models, and more data is needed for verification and analysis.