交通部運輸研究所自民國69年起，已辦理30期的「臺灣地區易肇事路段改善計畫」，建立了國內執行易肇事地點改善的基本作業程序與方法，本研究即針對易肇事路段改善計畫之執行績效進行評估研究。資料包絡分析法在國內、外文獻中多應用於效率分析，追溯過去運輸評價的文獻可發現，大多為效能分析且多應用經濟分析法。本研究應用資料包絡分析法探討易肇事地點改善的執行效能，以麥氏生產力指數探討各年的生產力變化，以及利用Tobit迴歸分析探討改善效能與道路環境因素及計畫相關因素之關係。研究結果發現，改善效能於執行後第二年較第一年下滑，之後大致持平，並於第五年再度下滑；投入改善措施後第二年與第五年的技術皆有進步，於其他年則持平。

Since 1980, the Institute of Transportation, Ministry of Transportation and Communications (MOTC) has administered the annual hazardous location improvement program (HLIP) for 30 consecutive years. The foundational studies and operation procedures for the HLIP have been established. This study aims to evaluate the performance of the HLIP. Most of the prior studies on the evaluation of transportation projects focused on the effectiveness analysis and the most prevailing technique was economic analysis. This study applies Data Envelopment Analysis (DEA) and Malmquist Productivity Index (MPI) to the examination of the performance of the HLIP. A Tobit model was developed to estimate the linear relationship between the effectiveness and influence factors. The results of this study indicate that the effectiveness deteriorated in the second and the fifth year, however technical change progress in the second and the fifth year.