中國與東南亞國家，由於土地、勞力相對於其他國家便宜，吸引了大量外資的進駐，貿易量持續高度成長。亞太地區之貨運量占了世界總量的六成，而中國地區更是亞太區域中貨櫃交易數量最多的國家。大陸與臺灣因臺灣海峽分隔兩地，彼此間因同文同種，而存在著既競爭又合作的特殊關係，因此此地區的海運市場相當值得我們去深入探討。本研究所採用的評估方式為資料包絡分析（Data Envelopment Analysis；DEA）之變形方法；資料包絡分析（Data Envelopment Analysis；DEA）為一種非參數分析法，主要使用數學規劃分析的技巧，有助於本研究找兩岸三地貨櫃港埠客觀資料的效率前緣，根據效率前緣來衡量評估兩岸三地貨櫃港生產效率，本研究以資料包絡分析之改進方法，遞迴資料包絡法（Recursive Data Envelopment Analysis；RDEA）對兩岸三地貨櫃港埠近三年生產效率做比較，並加以整合排序分析，以明瞭兩岸三地在這三年間各個貨櫃港埠之生產效率之消長，以作為其他研究之參考。

　　　　　In this article, we are going to apply a new procedure based on DEA (Data Envelopment Analysis) called cross time RDEA (Recursive Data Envelopment Analysis) to rank selected container ports in China and Taiwan region. The cross time RDEA is a mathematical programming skill for measuring the relative performance of decision making units (DMUs). Basically, the cross time RDEA is a nonparametric approach that does not require any assumption about the function form of the production function. However, it is controversy to rank DMUs based on their relative efficiencies, especially when their efficiencies are less than one. In this article, we will compare the ranks of ports obtained by three years’ DEA and RDEA respectively and drill down the result to unveil tactical moves needed to improve their efficiencies.