Since long ago, the whole performance of the Taiwan¡¬s airports presents the growth tendency besides the Cargo tonnes, the passengers and aircraft movements all present the decline tendency. Compares airport extension project which Taiwan continues to carry on every year, this is implying the performance of the Taiwan¡¬s airports presents bad. This research from 2003 to 2005 the Taiwan¡¬s 18 airports as the object of study, used Two-Stage DEA to do for the main methodology, and auxiliary by Malmquist index, Mann-Whitney test, and the Kruskal-Wallis test, carried on the performance measurement of the Taiwan¡¬s airports. This research divided the airport production process into two stages (include the production interface and the sales interface), so as to set an integral model. According to the analysis, the findings showed that the Taiwan¡¬s airports, had good performance, included C.K.S. international airport, Taipei Songshan airport, Magong airport, Chiayi airport, and Kinmen airport; the Taiwan¡¬s airports, presented the growth tendency from 2003 to 2005, included C.K.S. international airport, Tainan airport, Magong airport, Nangan airport, and Lanyu airport. The site characteristics, including airport class, the airport nature, the airport position, and airport location, have the significance influence regarding the performance of the Taiwan¡¬s airports.

傳統DEA效率分析假設產業具單一生產過程，直到2000年，Färe and Grosskopf 提出Network DEA，闡明產業生產過程應分屬多階段性質。本研究應用其架構，假設職棒產業生產過程為兩個階段，並特別以Sexton and Lewis (2003) 增加中間產出的Two-Stage DEA 法，即第一階段的產出作為第二階段的投入，進行1992年至2004年「中華職業棒球聯盟」，共71個DMUs的實證分析。 Two-Stage DEA的第一階段是指花錢聘雇球員，而球團有效花錢聘雇球員的程度稱為「前置效率」，效率平均值為0.994；第二階段效率是指球隊正式比賽時，球員是否充分發揮技術潛能贏球，稱為「臨場效率」，效率平均值為0.969；而包含第一、二階段的整體球團運作效率則為「組織效率」，效率平均值為0.798。透過與傳統DEA的BCC模型之比較，發現Two-Stage DEA提供較豐富的組織運作過程資訊，俾管理者找出球團之無效率階段。 復以Tobit截斷迴歸模型，探討影響球團「前置效率」、「臨場效率」與「組織效率」之變數為何。結果指出，球隊對戰觀眾數、聯盟變革與現場直播對「前置效率」有顯著影響，「臨場效率」則受到臨時性獎勵制度與投手平均年齡的影響，而對戰觀眾數、聯盟變革、總教練的更動頻率與常設性獎勵制度，則是造成球隊「組織效率」差異的主因。 關鍵字：Network DEA、中間產出、Two-Stage DEA、職業棒球、效率 / Traditional DEA gauges efficiencies with only one production process, while in this study we apply Network DEA initiated by Färe and Grosskopf (2000), and in particular follow the Two-Stage DEA model incorporating the intermediate products, outputs from the first stage becomes inputs to the second stage, by Sexton and Lewis (2003) to evaluate the production efficiency of 71 DMUs of the Chinese professional Baseball League (CPBL ) from 1992 to 2004. How fair are the ball teams paying the players is called ”front office efficiency”, arithmetic mean is 0.994, in the first stage, how potentially successful are the teams playing the games is called “on-field efficiency”, arithmetic mean is 0.969, in the second stage, and how potentially successful are the teams playing the games if with perfect front office efficiency of the teams is called “organization efficiency”, arithmetic mean is 0.798. Comparing Two-Stage DEA model with BCC model of traditional DEA, we find that the former model provides more information of organizational operations for managers to understand and better the performance of the teams. Tobit regression analysis shows that (1) the front office efficiency is significantly positively influenced by spectators, variation of the league and television live, and (2) the on-field efficiency is significantly positively influenced by extemporaneous bonus and pitcher’s age. (3) The organization efficiency is significantly positively influenced by the spectators, variation of the league and fixed bonus, but the organization efficiency is significantly negatively influenced by change of coaches. Keyword：Network DEA, Intermediate products, Two-Stage DEA , Efficiency, Professional baseball

This study adopt input orientation data envelopment Analysis approaches to evaluate the operational efficiency for major international airports in terms of the top 20 ranking by total passengers handling in Asia region in 2008. Unlike the most the other airport literature researches only using the one-stage DEA to measure the efficiency, in this study, two-stage, the extension of DEA, adding the concept of the intermediates product, separate the airport operation into airport capital side and airport capability side, is applied to evaluate the two-side efficiency of each airport under the assumption of constant returns to scale, variable returns to scale, scale efficiency and super efficiency model by employing four inputs, four intermediate products and two outputs. The result of this study shows that the China Shenzhen Baoan airport is the best airport in overall technical efficiency in different stages. It indicates that the operation of small size airports (category by the passengers handling) are run more efficiency than the large one. In addition, by two-stage DEA analysis, the airport capital usage is better than in capacity utilization. It found that the China Beijing Capital airport is good in airport infrastructure usage but the capacity utilization is not enough. Finally, the number of increasing return scale airports are more than the decreasing return scale airports among Asia region (e.g. Taiwan Taoyuan airport), therefore, the expansions of facility and capacity in those increasing return scale airports are worthy to invest to promote the operational efficiency and its competitiveness.