由於油價持續上漲，原物料成本增加與市場激烈競爭，公路客運業者面臨著相當大的衝擊。政府為改善大眾運輸績效與其經營環境，遂執行大眾運輸補貼作業。在資源有限的情況下，如何合理地將補貼額分配至相對具有效率的路線，儼然已成為政府當前亟待解決之課題。本研究之目的係以“偏遠路線＂之概念，將現有公路客運受補貼路線進行分群，並分析各集群路線間之一營運績效差異，以探討現有虧損補貼之適切性。本研究以99年度之營運資料為基礎，以空間資訊科學與集群分析進行偏遠路線之分類，並運用資料包絡分析法計算各集群路線營運績效。最後，本研究利用所建構之路線轉型策略矩陣探究各集群管理政策意涵，並提出後續研究之建議。

With the increased oil price, high costs of raw materials and strong competition in the market, unparalleled challenge faced the regional bus services. In order to improve the operation performance and environment of public transit, government has conducted the subsidization of public transportation. Under the constraint of limited resources, it is an important issue that how to allocate the subsidy to the relative efficiency routes rationally. The purpose of this study is to categorize the subsidy routes based on the concept of “rural route”, and evaluate the operation performance for each group in order to discuss the appropriateness of existing subsidy. This study uses the real data of subsidized regional bus service in 2010. Geospatial information science and cluster analysis were applied to categorize the subsidy routes. And DEA is utilized to evaluate the operation performance of all the routes in each group. Finally, this study addresses the managerial implications based on a matrix of route transformation proposed by this study, and offers suggestions to the future research.